2024/02/02 更新

写真b

ハン シェンファー
韓 先花
HAN Xianhua
*大学が定期的に情報更新している項目(その他は、researchmapの登録情報を転載)
所属*
理学部 数学科
人工知能科学研究科 人工知能科学専攻 博士課程前期課程
人工知能科学研究科 人工知能科学専攻 博士課程後期課程
職名*
教授
学位
工学博士 ( 琉球大学 )
研究キーワード
  • 機械学習、パターン認識、超解像度技術、画像理解・認識、コンピュタービジョン

  • 学内職務経歴*
    • 2023年9月 - 現在 
      理学部   数学科   教授
    • 2023年9月 - 現在 
      人工知能科学研究科   人工知能科学専攻 博士課程前期課程   教授
    • 2023年9月 - 現在 
      人工知能科学研究科   人工知能科学専攻 博士課程後期課程   教授
     

    研究分野

    • 情報通信 / 知能情報学

    • 情報通信 / 知覚情報処理

    経歴

    • 2023年9月 - 現在 
      立教大学   人工知能科学研究科   教授

      詳細を見る

    • 2017年3月 - 2023年8月 
      山口大学   大学院創成科学研究科   准教授

      詳細を見る

    • 2016年4月 - 2017年2月 
      産業技術総合研究所   人工知能研究センター   主任研究員

      詳細を見る

    • 2013年4月 - 2016年3月 
      立命館大学 グローバルイノベーション研究機構 研究教員(准教授)

      詳細を見る

    • 2010年4月 - 2013年3月 
      立命館大学 グローバルイノベーション研究機構 ポストドクトラルフェロー

      詳細を見る

    • 2007年4月 - 2010年3月 
      立命館大学 総合理工研究機構 ポストドクトラルフェロー

      詳細を見る

    • 2006年2月 - 2007年3月 
      中国中南林業大学 情報学科 講師

      詳細を見る

    ▼全件表示

    学歴

    • 2002年10月 - 2005年9月 
      琉球大学大学院

      詳細を見る

    • 1999年9月 - 2002年7月 
      中国山東大学大学院

      詳細を見る

    • 1995年9月 - 1999年7月 
      中国重慶大学

      詳細を見る

    受賞

    • 2012年5月  
      日本私立学校振興・共済事業団  学術研究振興基金(若手研究者奨励金) 

      詳細を見る

    論文

    • A Lightweight Network for Contextual and Morphological Awareness for Hepatic Vein Segmentation. 査読有り 国際共著 国際誌

      Guoyu Tong, Huiyan Jiang, Tianyu Shi, Xian-Hua Han, Yu-Dong Yao

      IEEE Journal of Biomedical and Health Informatics27 ( 10 ) 4878 - 4889   2023年10月

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/JBHI.2023.3305644

      researchmap

    • 3D Multiple-Contextual ROI-Attention Network for Efficient and Accurate Volumetric Medical Image Segmentation. 査読有り

      He Li, Yutaro Iwamoto, Xianhua Han, Lanfen Lin, Akira Furukawa, Shuzo Kanasaki, Yen-Wei Chen 0001

      IEICE Transactions on Information & Systems106 ( 5 ) 1027 - 1037   2023年5月

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1587/transinf.2022edp7193

      researchmap

    • IDH mutation status prediction by a radiomics associated modality attention network. 査読有り 国際共著 国際誌

      Xinran Zhang, Xiaoyu Shi, Yutaro Iwamoto, Jingliang Cheng, Jie Bai, Guohua Zhao, Xian-Hua Han, Yen-Wei Chen 0001

      The Visual Computer39 ( 6 ) 2367 - 2379   2023年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1007/s00371-022-02452-y

      researchmap

    • Exploring Cross Modality Feature Fusion for Activity Recognition at OpenPack Challenge 2022. 査読有り

      Tetsuo Inoshita, Yuto Namba, Yuichi Nakatani, Kenta Ishihara, Sachio Iwasaki, Kosuke Moriwaki, Xian-Hua Han

      PerCom Workshops   262 - 263   2023年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/PerComWorkshops56833.2023.10150371

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/percom/percomw2023.html#InoshitaNNIIMH23

    • Bottleneck Transformer model with Channel Self-Attention for skin lesion classification. 査読有り

      Masato Tada, Xian-Hua Han

      MVA   1 - 5   2023年

      詳細を見る

      担当区分:最終著者, 責任著者   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.23919/MVA57639.2023.10215720

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mva/mva2023.html#TadaH23

    • Investigating self-supervised learning for Skin Lesion Classification. 査読有り

      Takumi Morita, Xian-Hua Han

      MVA   1 - 5   2023年

      詳細を見る

      担当区分:最終著者   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.23919/MVA57639.2023.10215580

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mva/mva2023.html#MoritaH23

    • A Hybrid Wheat Head Detection model with Incorporated CNN and Transformer. 査読有り

      Sho Harada, Xian-Hua Han

      MVA   1 - 5   2023年

      詳細を見る

      担当区分:最終著者   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.23919/MVA57639.2023.10216087

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mva/mva2023.html#HaradaH23

    • Zero-Shot Blind Learning for Single-Image Super-Resolution. 査読有り

      Kazuhiro Yamawaki, Xian-Hua Han

      Inf.14 ( 1 ) 33 - 33   2023年1月

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(学術雑誌)  

      DOI: 10.3390/info14010033

      researchmap

    • Multi-Scale Channel Transformer Network for Single Image Deraining. 査読有り

      Yuto Namba, Xian-Hua Han

      MMAsia   20 - 7   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1145/3551626.3564946

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mmasia/mmasia2022.html#NambaH22

    • Deep Image and Kernel Prior Learning for Blind Super-Resolution. 査読有り

      Kazuhiro Yamawaki, Xian-Hua Han

      MMAsia   2 - 7   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1145/3551626.3564958

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mmasia/mmasia2022.html#YamawakiH22

    • Deep Unsupervised Blind Learning for Single Image Super Resolution. 査読有り

      Kazuhiro Yamawaki, Xian-Hua Han

      MIPR   190 - 193   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/MIPR54900.2022.00040

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mipr/mipr2022.html#YamawakiH22

    • An Accurate Unsupervised Liver Lesion Detection Method Using Pseudo-lesions. 査読有り 国際共著

      He Li, Yutaro Iwamoto, Xianhua Han, Lanfen Lin, Hongjie Hu, Yen-Wei Chen 0001

      MICCAI (8)   214 - 223   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1007/978-3-031-16452-1_21

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/miccai/miccai2022-8.html#LiIHLHC22

    • ScaleFormer: Revisiting the Transformer-based Backbones from a Scale-wise Perspective for Medical Image Segmentation. 査読有り 国際共著

      Huimin Huang, Shiao Xie, Lanfen Lin, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen 0001, Ruofeng Tong 0001

      IJCAI   964 - 971   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.24963/ijcai.2022/135

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/ijcai/ijcai2022.html#HuangXLIH0022

    • Generalized Deep Internal Learning for Hyperspectral Image Super Resolution. 査読有り

      Zhe Liu 0039, Xian-Hua Han

      ICIP   2641 - 2645   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICIP46576.2022.9897182

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icip/icip2022.html#0039H22

    • Unsupervised Generative Network for Blind Hyperspectral Image Super-Resolution. 査読有り 国際共著

      Zhe Liu 0039, Xianhua Han, Jiande Sun 0001, Yen-Wei Chen 0001

      ICIP   2621 - 2625   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICIP46576.2022.9897424

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icip/icip2022.html#0039H0022

    • Hyperspectral Reconstruction Using Auxiliary Rgb Learning From A Snapshot Image. 査読有り

      Kazuhiro Yamawaki, Yorimoto Kohei, Xian-Hua Han

      ICIP   186 - 190   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICIP46576.2022.9897926

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icip/icip2022.html#YamawakiKH22

    • Deep Learning-based Risk Prediction Model for Recurrence-free Survival in Patients with Hepatocellular Carcinoma Using Multi-phase CT Image. 査読有り 国際共著

      Weibin Wang, Fang Wang, Yunjun Yang, Yinhao Li, Jing Liu 0041, Xianhua Han, Lanfen Lin, Ruofeng Tong 0001, Hongjie Hu, Yen-Wei Chen 0001

      GCCE   926 - 929   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/GCCE56475.2022.10014204

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/gcce/gcce2022.html#WangWYLLHLTHC22

    • Synthesizing Contrast-enhanced Computed Tomography Images with an Improved Conditional Generative Adversarial Network.

      Yulin Yang, Yutaro Iwamoto, Yen-Wei Chen 0001, Caie Xu, Qingqing Chen, Hongjie Hu, Xian-Hua Han, Ruofeng Tong 0001, Lanfen Lin

      EMBC   2097 - 2100   2022年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/EMBC48229.2022.9871672

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/embc/embc2022.html#YangICXCHHTL22

    • Dual Discriminator-Based Unsupervised Domain Adaptation Using Adversarial Learning for Liver Segmentation on Multiphase CT Images. 査読有り 国際共著

      Swathi Ananda, Yutaro Iwamoto, Xianhua Han, Lanfen Lin, Hongjie Hu, Yen-Wei Chen 0001

      EMBC   1552 - 1555   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/EMBC48229.2022.9871188

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/embc/embc2022.html#AnandaIHLHC22

    • Unsupervised Domain Adaptation Using Adversarial Learning and Maximum Square Loss for Liver Tumors Detection in Multi-phase CT Images. 査読有り 国際共著

      Rahul Kumar Jain, Takahiro Sato, Taro Watasue, Tomohiro Nakagawa, Yutaro Iwamoto, Xianhua Han, Lanfen Lin, Hongjie Hu, Xiang Ruan, Yen-Wei Chen 0001

      EMBC   1536 - 1539   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/EMBC48229.2022.9871539

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/embc/embc2022.html#JainSWNIHLHRC22

    • Residual Multilayer Perceptrons for Genotype-Guided Recurrence Prediction of Non-Small Cell Lung Cancer. 査読有り

      Yang Ai, Panyanat Aonpong, Weibin Wang, Yinhao Li, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen 0001

      EMBC   447 - 450   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/EMBC48229.2022.9871896

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/embc/embc2022.html#AiAWLIHC22

    • Dual Heterogeneous Complementary Networks for Single Image Deraining. 査読有り

      Yuuto Nanba, Hikaru Miyata, Xian-Hua Han

      CVPR Workshops   567 - 576   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/CVPRW56347.2022.00072

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/cvpr/cvpr2022w.html#NanbaMH22

    • MTL-ABS3Net: Atlas-Based Semi-Supervised Organ Segmentation Network With Multi-Task Learning for Medical Images. 査読有り 国際共著 国際誌

      Huimin Huang, Qingqing Chen, Lanfen Lin, Ming Cai, Qiaowei Zhang, Yutaro Iwamoto, Xianhua Han, Akira Furukawa, Shuzo Kanasaki, Yen-Wei Chen 0001, Ruofeng Tong 0001, Hongjie Hu

      IEEE Journal of Biomedical and Health Informatics26 ( 8 ) 3988 - 3998   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/JBHI.2022.3153406

      researchmap

    • DPE-BoTNeT: Dual Position Encoding Bottleneck Transformer Network for Skin Lesion Classification. 査読有り

      Katsuhiro Nakai, Xian-Hua Han

      19th IEEE International Symposium on Biomedical Imaging(ISBI)   1 - 5   2022年

      詳細を見る

      担当区分:最終著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ISBI52829.2022.9761578

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/isbi/isbi2022.html#NakaiH22

    • A Weakly-Supervised Anomaly Detection Method via Adversarial Training for Medical Images. 査読有り 国際共著

      He Li, Yutaro Iwamoto, Xianhua Han, Lanfen Lin, Ruofeng Tong 0001, Hongjie Hu, Akira Furukawa, Shuzo Kanasaki, Yen-Wei Chen 0001

      IEEE International Conference on Consumer Electronics(ICCE)   1 - 4   2022年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ICCE53296.2022.9730129

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/iccel/icce2022.html#LiIHLTHFKC22

    • Mixed Transformer U-Net for Medical Image Segmentation. 査読有り 国際共著

      Hongyi Wang, Shiao Xie, Lanfen Lin, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen 0001, Ruofeng Tong 0001

      ICASSP   2390 - 2394   2022年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICASSP43922.2022.9746172

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icassp/icassp2022.html#WangXLIHCT22

    • Hyperspectral Image Reconstruction Using Multi-scale Fusion Learning. 国際誌

      Xian-Hua Han, Yinqiang Zheng, Yen-Wei Chen 0001

      ACM Transactions on Multimedia Computing, Communications, and Applications18 ( 1 ) 16 - 21   2022年

      詳細を見る

      担当区分:筆頭著者, 責任著者   掲載種別:研究論文(学術雑誌)  

      DOI: 10.1145/3477396

      researchmap

    • Mutual Information-Based Graph Co-Attention Networks for Multimodal Prior-Guided Magnetic Resonance Imaging Segmentation. 査読有り 国際共著 国際誌

      Shaocong Mo, Ming Cai, Lanfen Lin, Ruofeng Tong 0001, Qingqing Chen, Fang Wang, Hongjie Hu, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen 0001

      IEEE Transactions on Circuits and Systems for Video Technology32 ( 5 ) 2512 - 2526   2022年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/TCSVT.2021.3112551

      researchmap

    • EFNet: evidence fusion network for tumor segmentation from PET-CT volumes. 査読有り 国際共著 国際誌

      Zhaoshuo Diao, Huiyan Jiang, Xian-Hua Han, Yu-Dong Yao, Tianyu Shi

      Physics in medicine and biology66 ( 20 )   2021年10月8日

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)  

      Precise delineation of target tumor from positron emission tomography-computed tomography (PET-CT) is a key step in clinical practice and radiation therapy. PET-CT co-segmentation actually uses the complementary information of two modalities to reduce the uncertainty of single-modal segmentation, so as to obtain more accurate segmentation results. At present, the PET-CT segmentation methods based on fully convolutional neural network (FCN) mainly adopt image fusion and feature fusion. The current fusion strategies do not consider the uncertainty of multi-modal segmentation and complex feature fusion consumes more computing resources, especially when dealing with 3D volumes. In this work, we analyze the PET-CT co-segmentation from the perspective of uncertainty, and propose evidence fusion network (EFNet). The network respectively outputs PET result and CT result containing uncertainty by proposed evidence loss, which are used as PET evidence and CT evidence. Then we use evidence fusion to reduce uncertainty of single-modal evidence. The final segmentation result is obtained based on evidence fusion of PET evidence and CT evidence. EFNet uses the basic 3D U-Net as backbone and only uses simple unidirectional feature fusion. In addition, EFNet can separately train and predict PET evidence and CT evidence, without the need for parallel training of two branch networks. We do experiments on the soft-tissue-sarcomas and lymphoma datasets. Compared with 3D U-Net, our proposed method improves the Dice by 8% and 5% respectively. Compared with the complex feature fusion method, our proposed method improves the Dice by 7% and 2% respectively. Our results show that in PET-CT segmentation methods based on FCN, by outputting uncertainty evidence and evidence fusion, the network can be simplified and the segmentation results can be improved.

      DOI: 10.1088/1361-6560/ac299a

      PubMed

      researchmap

    • Multi-Scale Context Interaction Learning network for Medical Image Segmentation. 査読有り 国際共著 国際誌

      Wenhao Fang, Xian-Hua Han, Xu Qiao, Huiyan Jiang, Yen-Wei Chen 0001

      4th IEEE International Conference on Multimedia Information Processing and Retrieval(MIPR)   192 - 198   2021年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/MIPR51284.2021.00036

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mipr/mipr2021.html#FangHQJC21

    • 3D Graph-S2Net: Shape-Aware Self-ensembling Network for Semi-supervised Segmentation with Bilateral Graph Convolution. 査読有り 国際共著 国際誌

      Huimin Huang, Nan Zhou, Lanfen Lin, Hongjie Hu, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen 0001, Ruofeng Tong 0001

      Medical Image Computing and Computer Assisted Intervention - MICCAI 2021 - 24th International Conference   416 - 427   2021年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-87196-3_39

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/miccai/miccai2021-2.html#HuangZLHIHCT21

    • Deep Blind Un-Supervised Learning Network for Single Image Super Resolution. 査読有り

      Kazuhiro Yamawaki, Xian-Hua Han

      ICIP   1789 - 1793   2021年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICIP42928.2021.9506783

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icip/icip2021.html#YamawakiH21

    • HyperMixNet: Hyperspectral Image Reconstruction with Deep Mixed Network from a Snapshot Measurement. 査読有り

      Yorimoto Kohei, Xian-Hua Han

      ICCVW   1184 - 1193   2021年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICCVW54120.2021.00138

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/iccvw/iccvw2021.html#KoheiH21

    • Graph-Based Pyramid Global Context Reasoning With a Saliency- Aware Projection for Covid-19 Lung Infections Segmentation. 査読有り 国際共著

      Huimin Huang, Ming Cai, Lanfen Lin, Jing Zheng, Xiongwei Mao, Xiaohan Qian, Zhiyi Peng, Jianying Zhou 0006, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen 0001, Ruofeng Tong 0001

      IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP)   1050 - 1054   2021年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ICASSP39728.2021.9413957

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icassp/icassp2021.html#HuangCLZMQP0IH021

    • Improved Genotype-Guided Deep Radiomics Signatures for Recurrence Prediction of Non-Small Cell Lung Cancer. 査読有り 国際共著

      Panyanat Aonpong, Yutaro Iwamoto, Xian-Hua Han, Lanfen Lin, Yen-Wei Chen 0001

      EMBC   3561 - 3564   2021年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/EMBC46164.2021.9630703

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/embc/embc2021.html#AonpongIHLC21

    • An Efficient and Accurate 3D Multiple-Contextual Semantic Segmentation Network for Medical Volumetric Images. 査読有り

      He Li, Yutaro Iwamoto, Xianhua Han, Akira Furukawa, Shuzo Kanasaki, Yen-Wei Chen 0001

      EMBC   3309 - 3312   2021年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/EMBC46164.2021.9629671

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/embc/embc2021.html#LiIHFKC21

    • Medical Image Segmentation With Deep Atlas Prior. 査読有り 国際共著 国際誌

      Huimin Huang, Han Zheng, Lanfen Lin, Ming Cai, Hongjie Hu, Qiaowei Zhang, Qingqing Chen, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen 0001, Ruofeng Tong 0001

      IEEE Transactions on Medical Imaging40 ( 12 ) 3519 - 3530   2021年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/TMI.2021.3089661

      researchmap

    • Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions. 査読有り 国際誌

      Kohei Yamamichi, Xian-Hua Han

      IEEE Access9   146948 - 146958   2021年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/ACCESS.2021.3122450

      researchmap

    • Genotype-Guided Radiomics Signatures for Recurrence Prediction of Non-Small Cell Lung Cancer. 査読有り 国際共著 国際誌

      Panyanat Aonpong, Yutaro Iwamoto, Xian-Hua Han, Lanfen Lin, Yen-Wei Chen 0001

      IEEE Access9   90244 - 90254   2021年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/ACCESS.2021.3088234

      researchmap

    • Angular Margin Constrained Loss for Automatic Liver Fibrosis Staging. 査読有り 国際共著

      Katsuhiro Nakai, Xu Qiao, Xian-Hua Han

      MVA2021   1 - 5   2021年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.23919/MVA51890.2021.9511356

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mva/mva2021.html#NakaiQH21

    • A Tensor Sparse Representation-Based CBMIR System for Computer-Aided Diagnosis of Focal Liver Lesions and its Pilot Trial. 査読有り 国際共著

      Jian Wang, Xian-Hua Han, Lanfen Lin, Hongjie Hu, Yen-Wei Chen 0001

      ICMR2021   660 - 666   2021年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:ACM  

      DOI: 10.1145/3460426.3463673

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/mir/icmr2021.html#WangHLH021

    • Patch-Free 3D Medical Image Segmentation Driven by Super-Resolution Technique and Self-Supervised Guidance. 査読有り 国際共著

      Hongyi Wang, Lanfen Lin, Hongjie Hu, Qingqing Chen, Yinhao Li, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen 0001, Ruofeng Tong 0001

      Medical Image Computing and Computer Assisted Intervention – MICCAI 2021   131 - 141   2021年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-87193-2_13

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/miccai/miccai2021-1.html#WangLHCLIHCT21

    • Genotype-Guided Radiomics Signatures for Recurrence Prediction of Non-Small-Cell Lung Cancer. 国際共著

      Panyanat Aonpong, Yutaro Iwamoto, Xian-Hua Han, Lanfen Lin, Yen-Wei Chen 0001

      CoRRabs/2104.14420   2021年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/corr/corr2104.html#abs-2104-14420

    • Graph-based Pyramid Global Context Reasoning with a Saliency-aware Projection for COVID-19 Lung Infections Segmentation. 国際共著

      Huimin Huang, Ming Cai, Lanfen Lin, Jing Zheng, Xiongwei Mao, Xiaohan Qian, Zhiyi Peng, Jianying Zhou, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen, Ruofeng Tong

      CoRRabs/2103.04235   2021年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/corr/corr2103.html#abs-2103-04235

    • PA-ResSeg: A Phase Attention Residual Network for Liver Tumor Segmentation from Multi-phase CT Images. 国際共著

      Yingying Xu, Ming Cai, Lanfen Lin, Yue Zhang, Hongjie Hu, Zhiyi Peng, Qiaowei Zhang, Qingqing Chen, Xiongwei Mao, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen 0001, Ruofeng Tong 0001

      CoRRabs/2103.00274   2021年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/corr/corr2103.html#abs-2103-00274

    • Automatic Detection and Segmentation of Liver Tumors in Multi- phase CT Images by Phase Attention Mask R-CNN 査読有り 国際共著

      ICCE 2021   1 - 5   2021年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICCE50685.2021.9427760

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/iccel/icce2021.html#HasegawaIHLHC021

    • A Cascade of 2.5D CNN and Bidirectional CLSTM Network for Mitotic Cell Detection in 4D Microscopy Image. 査読有り 国際共著 国際誌

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Satoko Takemoto, Hideo Yokota, Sari Ipponjima, Tomomi Nemoto, Wei Xiong 0001, Yen-Wei Chen 0001

      IEEE ACM Trans. Comput. Biol. Bioinform.18 ( 2 ) 396 - 404   2021年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/TCBB.2019.2919015

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/tcbb/tcbb18.html#Kitrungrotsakul21

    • Deep Unsupervised Fusion Learning for Hyperspectral Image Super Resolution. 査読有り 国際誌

      Zhe Liu, Yinqiang Zheng, Xian-Hua Han

      Sensors21 ( 7 ) 2348 - 2348   2021年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(学術雑誌)  

      DOI: 10.3390/s21072348

      researchmap

    • MCGKT-Net: Multi-level Context Gating Knowledge Transfer Network for Single Image Deraining.

      Kohei Yamamichi, Xian-Hua Han

      CoRRabs/2010.09241   2020年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/corr/corr2010.html#abs-2010-09241

    • Multimodal Priors Guided Segmentation of Liver Lesions in MRI Using Mutual Information Based Graph Co-Attention Networks. 査読有り 国際共著

      Shaocong Mo, Ming Cai, Lanfen Lin, Ruofeng Tong, Qingqing Chen, Fang Wang, Hongjie Hu, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen

      MICCAI 2020: Medical Image Computing and Computer Assisted Intervention   429 - 438   2020年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-59719-1_42

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/miccai/miccai2020-4.html#MoCLTCWHIHC20

    • Deep Residual Attention Network for Hyperspectral Image Reconstruction. 査読有り

      Yorimoto Kohei, Xian-Hua Han

      ICPR2020   8547 - 8553   2020年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ICPR48806.2021.9412321

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icpr/icpr2020.html#KoheiH20

    • MCGKT-Net: Multi-level Context Gating Knowledge Transfer Network for Single Image Deraining. 査読有り

      Kohei Yamamichi, Xian-Hua Han

      ACCV 2020   68 - 83   2020年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-69532-3_5

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/accv/accv2020-2.html#YamamichiH20

    • Cell Detection and Segmentation in Microscopy Images with Improved Mask R-CNN. 査読有り

      Seiya Fujita, Xian-Hua Han

      ACCV Workshops 2020   58 - 70   2020年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-69756-3_5

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/accv/accv2020w.html#FujitaH20

    • Unsupervised Multispectral and Hyperspectral Image Fusion with Deep Spatial and Spectral Priors. 査読有り

      Zhe Liu, Yinqiang Zheng, Xian-Hua Han

      ACCV Workshops 2020   31 - 45   2020年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-69756-3_3

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/accv/accv2020w.html#LiuZH20

    • Spatial and Channel Attention Modulated Network for Medical Image Segmentation. 査読有り

      Wenhao Fang, Xian-Hua Han

      Computer Vision - ACCV 2020 Workshops - 15th Asian Conference on Computer Vision   3 - 17   2020年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-69756-3_1

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/accv/accv2020w.html#FangH20

    • CasCRNN-GL-Net: cascaded convolutional and recurrent neural networks with global and local pathways for classification of focal liver lesions in multi-phase CT images. 査読有り 国際共著 国際誌

      Dong Liang, Yingying Xu, Lanfen Lin, Nan Zhou, Hongjie Hu, Qiaowei Zhang, Qingqing Chen, Xianhua Han, Yutaro Iwamoto, Yen-Wei Chen 0001

      Commun. Inf. Syst.20 ( 4 ) 415 - 442   2020年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.4310/cis.2020.v20.n4.a2

      researchmap

    • UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation. 国際共著

      Huimin Huang, Lanfen Lin, Ruofeng Tong, Hongjie Hu, Qiaowei Zhang, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen, Jian Wu

      CoRRabs/2004.08790   2020年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/corr/corr2004.html#abs-2004-08790

    • WNET: An End-to-End Atlas-Guided and Boundary-Enhanced Network for Medical Image Segmentation. 査読有り 国際共著

      Huimin Huang, Lanfen Lin, Ruofeng Tong, Hongjie Hu, Qiaowei Zhang, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen, Jian Wu

      17th IEEE International Symposium on Biomedical Imaging(ISBI)   763 - 766   2020年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ISBI45749.2020.9098654

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/isbi/isbi2020.html#HuangLTHZIHCW20

    • Deep Learning Method for Content-Based Retrieval of Focal Liver Lesions Using Multiphase Contrast-Enhanced Computer Tomography Images. 査読有り 国際共著

      Yusuke Yoshinobu, Yutaro Iwamoto, Xianhua Han, Lanfen Lin, Hongjie Hu, Qiaowei Zhang, Yen-Wei Chen

      2020 IEEE International Conference on Consumer Electronics (ICCE)(ICCE)   1 - 4   2020年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ICCE46568.2020.9043172

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/iccel/icce2020.html#YoshinobuIHLHZC20

    • UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation 査読有り 国際共著

      ICASSP 2020   1055 - 1059   2020年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)  

      DOI: 10.1109/ICASSP40776.2020.9053405

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icassp/icassp2020.html#HuangLTHZIHCW20

    • Hyperspectral Reconstruction with Redundant Camera Spectral Sensitivity Functions. 査読有り 国際共著 国際誌

      Xian-Hua Han, Yinqiang Zheng, Jiande Sun, Yen-Wei Chen

      ACM Trans. Multim. Comput. Commun. Appl.16 ( 2 ) 57 - 15   2020年

      詳細を見る

      担当区分:筆頭著者, 責任著者   掲載種別:研究論文(学術雑誌)  

      DOI: 10.1145/3386313

      researchmap

    • Semi-Supervised Learning for Semantic Segmentation of Emphysema With Partial Annotations. 査読有り 国際共著 国際誌

      Liying Peng, Lanfen Lin, Hongjie Hu, Yue Zhang, Huali Li, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen

      IEEE J. Biomed. Health Informatics24 ( 8 ) 2327 - 2336   2020年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/JBHI.2019.2963195

      researchmap

    • An end-to-end CNN and LSTM network with 3D anchors for mitotic cell detection in 4D microscopic images and its parallel implementation on multiple GPUs. 査読有り 国際共著 国際誌

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Satoko Takemoto, Hideo Yokota, Sari Ipponjima, Tomomi Nemoto, Wei Xiong 0001, Yen-Wei Chen 0001

      Neural Comput. Appl.32 ( 10 ) 5669 - 5679   2020年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1007/s00521-019-04374-8

      researchmap

    • Image super-resolution based on two-level residual learning CNN. 査読有り 国際共著 国際誌

      Min Gao, Xian-Hua Han, Jing Li 0046, Hui Ji, Huaxiang Zhang 0001, Jiande Sun 0001

      Multim. Tools Appl.79 ( 7-8 ) 4831 - 4846   2020年

      詳細を見る

      担当区分:責任著者   掲載種別:研究論文(学術雑誌)  

      DOI: 10.1007/s11042-018-6751-5

      researchmap

    • Tensor-based sparse representations of multi-phase medical images for classification of focal liver lesions. 査読有り 国際共著 国際誌

      Jian Wang 0004, Jing Li, Xian-Hua Han, Lanfen Lin, Hongjie Hu, Yingying Xu, Qingqing Chen, Yutaro Iwamoto, Yen-Wei Chen

      Pattern Recognit. Lett.130   207 - 215   2020年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1016/j.patrec.2019.01.001

      researchmap

    • Semi-supervised Segmentation of Liver Using Adversarial Learning with Deep Atlas Prior. 査読有り 国際共著

      Han Zheng,Lanfen Lin, Hongjie Hu, Qiaowei Zhang, Qingqing Chen, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen, Ruofeng Tong, Jian Wu

      Medical Image Computing and Computer Assisted Intervention - MICCAI 2019 - 22nd International Conference, Shenzhen, China, October 13-17, 2019, Proceedings, Part VI   148 - 156   2019年

      詳細を見る

      出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-32226-7_17

      researchmap

    • A Cascade Attention Network for Liver Lesion Classification in Weakly-Labeled Multi-phase CT Images. 査読有り 国際共著

      Xiao Chen, Lanfen Lin, Hongjie Hu, Qiaowei Zhang, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen, Ruofeng Tong, Jian Wu

      Domain Adaptation and Representation Transfer and Medical Image Learning with Less Labels and Imperfect Data - First MICCAI Workshop, DART 2019, and First International Workshop, MIL3ID 2019, Shenzhen, Held in Conjunction with MICCAI 2019, Shenzhen, China   129 - 138   2019年

      詳細を見る

      出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-33391-1_15

      researchmap

    • Deep Learning-Based Radiomics Models for Early Recurrence Prediction of Hepatocellular Carcinoma with Multi-phase CT Images and Clinical Data. 査読有り 国際共著

      Weibin Wang, Qingqing Chen, Yutaro Iwamoto, Xianhua Han, Qiaowei Zhang, Hongjie Hu, Lanfen Lin, Yen-Wei Chen

      41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019, Berlin, Germany, July 23-27, 2019   4881 - 4884   2019年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/EMBC.2019.8856356

      researchmap

      その他リンク: https://dblp.uni-trier.de/conf/embc/2019

    • Classification and Quantification of Emphysema Using a Multi-Scale Residual Network. 査読有り 国際共著 国際誌

      Liying Peng, Yen-Wei Chen, Lanfen Lin, Hongjie Hu, Huali Li, Qingqing Chen, Xiaoli Ling, Dan Wang, Xianhua Han, Yutaro Iwamoto

      IEEE J. Biomed. Health Informatics23 ( 6 ) 2526 - 2536   2019年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/JBHI.2018.2890045

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/titb/titb23.html#PengCLHLCLWHI19

    • Multi-Level and Multi-Scale Spatial and Spectral Fusion CNN for Hyperspectral Image Super-Resolution. 査読有り

      Xian-Hua Han, Yinqiang Zheng, Yen-Wei Chen

      2019 IEEE/CVF International Conference on Computer Vision Workshops   4330 - 4339   2019年

      詳細を見る

      担当区分:筆頭著者, 責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ICCVW.2019.00533

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/iccvw/iccvw2019.html#HanZC19

    • Residual Component Estimating CNN for Image Super-Resolution. 査読有り

      Xian-Hua Han, Yongqing Sun, Yen-Wei Chen

      Fifth IEEE International Conference on Multimedia Big Data(BigMM)   443 - 447   2019年

      詳細を見る

      担当区分:筆頭著者, 責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/BigMM.2019.00028

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/bigmm/bigmm2019.html#HanSC19

    • Deep Residual Network of Spectral and Spatial Fusion for Hyperspectral Image Super-Resolution. 査読有り

      Xian-Hua Han, Yen-Wei Chen

      Fifth IEEE International Conference on Multimedia Big Data(BigMM)   266 - 270   2019年

      詳細を見る

      担当区分:筆頭著者, 責任著者   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/BigMM.2019.00-13

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/bigmm/bigmm2019.html#HanC19

    • Spectral Representation via Data-Guided Sparsity for Hyperspectral Image Super-Resolution. 査読有り 国際共著 国際誌

      Xian-Hua Han, Yongqing Sun, Jian Wang 0004, Boxin Shi, Yinqiang Zheng, Yen-Wei Chen

      Sensors19 ( 24 ) 5401 - 5401   2019年

      詳細を見る

      担当区分:筆頭著者, 責任著者   掲載種別:研究論文(学術雑誌)  

      DOI: 10.3390/s19245401

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/sensors/sensors19.html#HanSWSZC19

    • Automatic Segmentation of the Paranasal Sinus from Computer Tomography Images Using a Probabilistic Atlas and a Fully Convolutional Network. 査読有り 国際誌

      Yutaro Iwamoto, Kun Xiong, Takahiro Kitamura, Xian-Hua Han, Naoki Matsushiro, Hiroshi Nishimura, Yen-Wei Chen

      Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference2019   2789 - 2792   2019年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In this paper, we present an automatic approach to paranasal sinus segmentation in computed tomography (CT) images. The proposed method combines a probabilistic atlas and a fully convolutional network (FCN). The probabilistic atlas was used to automatically localize the paranasal sinus and determine its bounding box. The FCN was then used to automatically segment the paranasal sinus in the bounding box. Comparing our proposed method with the conventional FCN (without probabilistic atlas) and the state-of-the-art method using active contour with group similarity, the proposed method demonstrated an improvement in the paranasal sinus segmentation. The segmentation accuracy (Dice coefficient) was about 0.83 even for the case with unclear boundary.

      DOI: 10.1109/EMBC.2019.8856703

      PubMed

      researchmap

      その他リンク: https://dblp.uni-trier.de/conf/embc/2019

    • VesselNet: A deep convolutional neural network with multi pathways for robust hepatic vessel segmentation. 査読有り 国際共著 国際誌

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Lanfen Lin, Amir Hossein Foruzan, Wei Xiong 0001, Yen-Wei Chen

      Comput. Medical Imaging Graph.75   74 - 83   2019年

      詳細を見る

    • A Cascade of CNN and LSTM Network with 3D Anchors for Mitotic Cell Detection in 4D Microscopic Image. 査読有り 国際共著

      Titinunt Kitrungrotsakul, Yutaro Iwamoto, Xian-Hua Han, Satoko Takemoto, Hideo Yokota, Sari Ipponjima, Tomomi Nemoto, Xiong Wei 0001, Yen-Wei Chen

      IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2019, Brighton, United Kingdom, May 12-17, 2019   1239 - 1243   2019年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/ICASSP.2019.8682326

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/conf/icassp/icassp2019.html#Kitrungrotsakul19

    • Adaptive Semi-Supervised Feature Selection for Cross-Modal Retrieval. 査読有り 国際共著 国際誌

      En Yu, Jiande Sun, Jing Li 0046, Xiaojun Chang, Xian-Hua Han, Alexander G. Hauptmann

      IEEE Trans. Multimedia21 ( 5 ) 1276 - 1288   2019年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1109/TMM.2018.2877127

      researchmap

    • Three-Dimensional Embryonic Image Segmentation and Registration Based on Shape Index and Ellipsoid-Fitting Method. 査読有り 国際共著 国際誌

      Sihai Yang, Xianhua Han, Yen-Wei Chen

      J. Comput. Biol.26 ( 2 ) 128 - 142   2019年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1089/cmb.2018.0165

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/jcb/jcb26.html#YangHC19

    • Automatic Liver Segmentation Using U-Net with Wasserstein GANs 国際誌

      Yuki Enokiya, Yutaro Iwamoto, Yen-Wei Chen, Xian-Hua Han

      Journal of Image and Graphics6 ( 2 ) 152-159   2018年12月

      詳細を見る

    • Generic and Specific Impressions Estimation and Their Application to KANSEI-Based Clothing Fabric Image Retrieval 査読有り 国際共著 国際誌

      Yen-Wei Chen, Xinyin Huang, Dingye Chen, Xian-Hua Han

      International Journal of Pattern Recognition and Artificial Intelligence32 ( 10 )   2018年10月1日

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:World Scientific Publishing Co. Pte Ltd  

      Current image retrieval techniques are mainly based on text or visual contents. However, both text-based and contents-based methods lack the capability of utilizing human intuition and KANSEI (impression). In this paper, we proposed an impression-based image retrieval method in order to realize the image retrieval according to our impression presented by impression keywords. We first propose a generic and specific impressions estimation method based on machine learning and then apply it to impression-based clothing fabric image retrieval. We use a semantic differential (SD) method to measure the user's impressions such as brightness and warmth while they view a cloth fabric image. We also extract both global and local features of cloth fabric images such as color and texture using computer vision techniques. Then we use support vector regression to model the mapping functions between the generic impression (or specific impression) and image features. The learnt mapping functions are used to estimate the generic and specific impressions of cloth fabric images. The retrieval is done by comparing the query impression with the estimated impression of images in the database.

      DOI: 10.1142/S0218001418540241

      Scopus

      researchmap

    • Joint weber-based rotation invariant uniform local ternary pattern for classification of pulmonary emphysema in CT images 査読有り 国際共著

      Liying Peng, Lanfen Lin, Hongjie Hu, Xiaoli Ling, Dan Wang, Xianhua Han, Yen-Wei Chen

      Proceedings - International Conference on Image Processing, ICIP2017-   2050 - 2054   2018年2月20日

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      In this paper, we present a novel image representation approach for classifying emphysema in computed tomography (CT) images of the lung. Our proposed method extends rotation invariant uniform local binary pattern (RIULBP) and local ternary pattern (LTP), which are extensively used in a variety of computer vision applications, into rotation invariant uniform local ternary pattern (RIULTP) with a human perception principle: Weber's law. In addition, by integrating the upper pattern and the lower pattern of the Weber-based RIULTP (WRIULTP), we further put forward the joint Weber-based rotation invariant uniform local ternary pattern (JWRIULTP), which allows for a much richer representation and also takes the comprehensive information of the image into account. The proposed methods are tested on the Outex database (texture database) and the Bruijne and Sorensen database (emphysema database). The results show the superiority of the proposed approaches to the state-of-the-art techniques for emphysema classification including rotation invariant local binary pattern (RILBP) and texton-based approach.

      DOI: 10.1109/ICIP.2017.8296642

      Scopus

      researchmap

    • High-order statistics of micro-texton for HEp-2 staining pattern classification 査読有り

      Xian-Hua Han, Yen-Wei Chen

      Intelligent Systems Reference Library140   135 - 164   2018年

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:論文集(書籍)内論文   出版者・発行元:Springer Science and Business Media Deutschland GmbH  

      This study addresses the classification problem of the HEp-2 cell using indirect immunofluorescent (IIF) image analysis, which can indicate the presence of autoimmune diseases by finding antibodies in the patient serum. Generally, the method used for IIF analysis remains subjective, and depends too heavily on the experience and expertise of the physician. Recently, studies have shown that it is possible to identify the cell patterns using IIF image analysis and machine learning techniques. However, it still has large gap in recognition rates to the physical experts’ one. This paper explores an approach in which the discriminative features of HEp-2 cell images in IIF are extracted and then, the patterns of the HEp-2 cell are identified using machine learning techniques. This study aims to realize a method for extracting highly-discriminant features from HEp-2 cell images by exploring a robust local descriptor inspired by Weber’s law. The investigated local descriptor is based on the fact that human perception for distinguishing a pattern depends not only on the absolute intensity of the stimulus but also on the relative variance of the stimulus. Therefore, we firstly transform the original stimulus (the images in our study) into a differential excitation-domain according to Weber’s law, and then explore a local patch, also called as micro-Texton, in the transformed domain as Weber local descriptor. Furthermore, we propose to employ a parametric probability process to model the Weber local descriptors, and extract the higher-order statistics to the model parameters for image representation. The proposed strategy can adaptively characterize the Weber local descriptor space using generative probability model, and then learn the parameters for better fitting the training space, which would lead to more discriminant representation for HEp-2 cell images. The simple linear support vector machine is used for cell pattern identification because of its low computational cost, in particular for large-scale datasets. Experiments using the open HEp-2 cell dataset used in the ICIP2013 contest validate that the proposed strategy can achieve a much better performance than the widely used local binary pattern (LBP) histogram and its extensions, Rotation Invariant Co-occurrence LBP (RICLBP) and Pairwise Rotation Invariant Co-occurrence LBP (PRICoLBP), and that the achieved recognition error rate is even very significantly below the observed intra-laboratory variability.

      DOI: 10.1007/978-3-319-68843-5_6

      Scopus

      researchmap

    • Automatic segmentation of cellular/nuclear boundaries based on the shape index of image intensity surfaces 査読有り

      Si-Hai Yang, Xian-Hua Han, Yen-Wei Chen

      Smart Innovation, Systems and Technologies71   67 - 77   2018年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer Science and Business Media Deutschland GmbH  

      Segmentation of cellular and nuclear boundaries in differential interference contrast microscopy images is an important pre-processing step for biological image analysis. It is considered as a challenging task because of the interference of cell walls, blurs, nonuniform intensity background, and poor contrast between the foreground and the background. In this paper, we present a novel scheme on cellular boundary segmentation. Based on shape index (SI), the proposed method focuses on the detection of cytoplasm granules inside cellular regions. With several geometric post-processing techniques, the SI thresholding results are integrated into the segmented images. Because the size of the cytoplasm granules is usually too small comparing with the thickness of focal planes in Z-stack, we can not calculate SI values according to the method of constructing the intensity isosurface in 3D images. Consequently, we regard intensity as Z coordinate and compute SI values within each slice. A computed SI represents the shape of intensity surface or the variation of intensity near to a target pixel. Furthermore, we also show the proposed method can be applied to nuclear segmentation with a different post-processing step. Experimental results show the proposed algorithm has higher accuracy than existing schemes despite the existence of cell walls with different shapes and fluctuated intensities.

      DOI: 10.1007/978-3-319-59397-5_8

      Scopus

      researchmap

    • Automatic and robust vessel segmentation in CT volumes using submodular constrained graph 査読有り

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Yen-Wei Chen

      Smart Innovation, Systems and Technologies71   57 - 66   2018年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer Science and Business Media Deutschland GmbH  

      Graph cut is one of segmentation method that can give us the good result on natural image and large organ segmentation of medical image. However, we cannot get the correct or accurate results by using graph cut on detailed structures such as, tree branch, or blood vessel because the property of smoothness in graph cuts energy function will completely remove the small branch of the detailed structure to minimize its cost. We propose the vessel extraction method which combine graph cut and concept of submodular function. The conventional graph cuts will be use to obtain initial segmentation while graph cut with submodular function will be use to refine the initial segmentation. Submodular function can solve the problem of smoothness of graph cut in detail structure as shown in result that less segment and more united vessel tree than conventional graph cuts. The experimental result shows that our method can segment blood vessels of liver with higher accuracy while graph cut lead to a lot of loss of the detail branches in the liver vessel. With submodular constraint, we can connect the segment branch of vessel into united vessel tree which conventional graph cut still remain the segment of vessel’s branches.

      DOI: 10.1007/978-3-319-59397-5_7

      Scopus

      researchmap

    • Detection of liver tumor candidates from CT images using deep convolutional neural networks 査読有り

      Yoshihiro Todoroki, Xian-Hua Han, Yutaro Iwamoto, Lanfen Lin, Hongjie Hu, Yen-Wei Chen

      Smart Innovation, Systems and Technologies71   140 - 145   2018年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer Science and Business Media Deutschland GmbH  

      There are multiple types of tumors occurring in the liver. Different tumors have different visual appearance and their visual appearance changes after injection of the contrast medium. So detection of liver tumors is considered as a challenging task. In this paper, we propose a method for detection of liver tumor candidates from CT images using a deep convolutional neural network. Experimental results show that we can significantly improve the detection accuracy by using our proposed method compared with the previous researches.

      DOI: 10.1007/978-3-319-59397-5_15

      Scopus

      researchmap

    • Self-Similarity Constrained Sparse Representation for Hyperspectral Image Super-Resolution. 査読有り 国際誌

      Xian-Hua Han, Boxin Shi, Yinqiang Zheng

      IEEE Trans. Image Processing27 ( 11 ) 5625 - 5637   2018年

      詳細を見る

      担当区分:筆頭著者  

      DOI: 10.1109/TIP.2018.2855418

      researchmap

    • Texture-specific bag of visual words model and spatial cone matching-based method for the retrieval of focal liver lesions using multiphase contrast-enhanced CT images 査読有り 国際共著 国際誌

      Yingying Xu, Lanfen Lin, Hongjie Hu, Dan Wang, Wenchao Zhu, Jian Wang, Xian-Hua Han, Yen-Wei Chen

      International Journal of Computer Assisted Radiology and Surgery13 ( 1 ) 151 - 164   2018年1月1日

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:Springer Verlag  

      Purpose: The bag of visual words (BoVW) model is a powerful tool for feature representation that can integrate various handcrafted features like intensity, texture, and spatial information. In this paper, we propose a novel BoVW-based method that incorporates texture and spatial information for the content-based image retrieval to assist radiologists in clinical diagnosis. Methods: This paper presents a texture-specific BoVW method to represent focal liver lesions (FLLs). Pixels in the region of interest (ROI) are classified into nine texture categories using the rotation-invariant uniform local binary pattern method. The BoVW-based features are calculated for each texture category. In addition, a spatial cone matching (SCM)-based representation strategy is proposed to describe the spatial information of the visual words in the ROI. In a pilot study, eight radiologists with different clinical experience performed diagnoses for 20 cases with and without the top six retrieved results. A total of 132 multiphase computed tomography volumes including five pathological types were collected. Results: The texture-specific BoVW was compared to other BoVW-based methods using the constructed dataset of FLLs. The results show that our proposed model outperforms the other three BoVW methods in discriminating different lesions. The SCM method, which adds spatial information to the orderless BoVW model, impacted the retrieval performance. In the pilot trial, the average diagnosis accuracy of the radiologists was improved from 66 to 80% using the retrieval system. Conclusion: The preliminary results indicate that the texture-specific features and the SCM-based BoVW features can effectively characterize various liver lesions. The retrieval system has the potential to improve the diagnostic accuracy and the confidence of the radiologists.

      DOI: 10.1007/s11548-017-1671-9

      Scopus

      researchmap

    • Multi-pathways CNN for robust vascular segmentation. 査読有り 国際共著

      Titinunt Kitrungrotsakul, Xian-Hua Han, Xiong Wei, Yen-Wei Chen

      Medical Imaging 2018: Biomedical Applications in Molecular, Structural, and Functional Imaging, Houston, Texas, United States, 10-15 February 2018   105781S   2018年

      詳細を見る

      出版者・発行元:SPIE  

      DOI: 10.1117/12.2293074

      researchmap

    • Comprehensive Study of Multiple CNNs Fusion for Fine-Grained Dog Breed Categorization. 査読有り

      Minori Uno, Xian-Hua Han, Yen-Wei Chen

      2018 IEEE International Symposium on Multimedia, ISM 2018, Taichung, Taiwan, December 10-12, 2018   198 - 203   2018年

      詳細を見る

      担当区分:責任著者   出版者・発行元:IEEE Computer Society  

      DOI: 10.1109/ISM.2018.000-7

      researchmap

    • Interactive Liver Segmentation in CT Volumes Using Fully Convolutional Networks. 査読有り 国際共著

      Titinunt Kitrungrotsakul, Yutaro Iwamoto, Xian-Hua Han, Wei Xiong, Lanfen Lin, Hongjie Hu, Huiyan Jiang, Yen-Wei Chen

      Intelligent Interactive Multimedia Systems and Services, Proceedings of 2018 Conference, KES IIMSS 2018, Gold Cost, Australia, 20-22 June 2018, Proceedings   216 - 222   2018年

      詳細を見る

      出版者・発行元:Springer  

      DOI: 10.1007/978-3-319-92231-7_22

      researchmap

    • Residual HSRCNN: Residual Hyper-Spectral Reconstruction CNN from an RGB Image. 査読有り

      Xian-Hua Han, Boxin Shi, Yinqiang Zheng

      24th International Conference on Pattern Recognition, ICPR 2018, Beijing, China, August 20-24, 2018   2664 - 2669   2018年

      詳細を見る

      担当区分:筆頭著者, 責任著者   出版者・発行元:IEEE Computer Society  

      DOI: 10.1109/ICPR.2018.8545634

      researchmap

    • A 2.5D Cascaded Convolutional Neural Network with Temporal Information for Automatic Mitotic Cell Detection in 4D Microscopic Images. 査読有り 国際共著

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Satoko Takemoto, Hideo Yokota, Sari Ipponjima, Tomomi Nemoto, Xiong Wei, Yen-Wei Chen

      14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2018, Huangshan, China, July 28-30, 2018   202 - 205   2018年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/FSKD.2018.8687125

      researchmap

    • Semi-Automatic Segmentation of Paranasal Sinus from CT images Using Fully Convolutional Networks. 査読有り

      Kun Xiong, Takahiro Kitamura, Yutaro Iwamoto, Xian-Hua Han, Naoki Matsushiro, Hiroshi Nishimura, Yen-Wei Chen

      IEEE 7th Global Conference on Consumer Electronics, GCCE 2018, Nara, Japan, October 9-12, 2018   268 - 269   2018年

      詳細を見る

      出版者・発行元:IEEE  

      DOI: 10.1109/GCCE.2018.8574753

      researchmap

    • Residual Convolutional Neural Networks with Global and Local Pathways for Classification of Focal Liver Lesions. 査読有り 国際共著

      Dong Liang, Lanfen Lin, Hongjie Hu, Qiaowei Zhang, Qingqing Chen, Yutaro lwamoto, Xianhua Han, Yen-Wei Chen

      PRICAI 2018: Trends in Artificial Intelligence - 15th Pacific Rim International Conference on Artificial Intelligence, Nanjing, China, August 28-31, 2018, Proceedings, Part I   617 - 628   2018年

      詳細を見る

      出版者・発行元:Springer  

      DOI: 10.1007/978-3-319-97304-3_47

      researchmap

    • Combining Convolutional and Recurrent Neural Networks for Classification of Focal Liver Lesions in Multi-phase CT Images. 査読有り 国際共著

      Dong Liang, Lanfen Lin, Hongjie Hu, Qiaowei Zhang, Qingqing Chen, Yutaro lwamoto, Xianhua Han, Yen-Wei Chen

      Medical Image Computing and Computer Assisted Intervention - MICCAI 2018 - 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part II   666 - 675   2018年

      詳細を見る

      出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-00934-2_74

      researchmap

    • Classification of Pulmonary Emphysema in CT Images Based on Multi-Scale Deep Convolutional Neural Networks. 査読有り 国際共著

      Liying Peng, Lanfen Lin, Hongjie Hu, Huali Li, Xiaoli Ling, Dan Wang, Xianhua Han, Yutaro Iwamoto, Yen-Wei Chen

      2018 IEEE International Conference on Image Processing, ICIP 2018, Athens, Greece, October 7-10, 2018   3119 - 3123   2018年

      詳細を見る

      出版者・発行元:IEEE  

      DOI: 10.1109/ICIP.2018.8451514

      researchmap

    • A Cascade of 2.5D CNN and LSTM Network for Mitotic Cell Detection in 4D Microscopy Image. 国際共著

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Satoko Takemoto, Hideo Yokota, Sari Ipponjima, Tomomi Nemoto, Wei Xion, Yen-Wei Chen

      CoRRabs/1806.01018   2018年

      詳細を見る

    • Focal Liver Lesion Classification Based on Tensor Sparse Representations of Multi-phase CT Images. 査読有り 国際共著

      Jian Wang, Xian-Hua Han, Jiande Sun, Lanfen Lin, Hongjie Hu, Yingying Xu, Qingqing Chen, Yen-Wei Chen

      Advances in Multimedia Information Processing - PCM 2018 - 19th Pacific-Rim Conference on Multimedia, Hefei, China, September 21-22, 2018, Proceedings, Part II   696 - 704   2018年

      詳細を見る

      出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-00767-6_64

      researchmap

    • Hyperspectral Image Classification Using Nonnegative Sparse Spectral Representation and Spatial Regularization. 査読有り 国際共著

      Xian-Hua Han, Jian Wang, JianDe Sun, Yen-Wei Chen

      Advances in Multimedia Information Processing - PCM 2018 - 19th Pacific-Rim Conference on Multimedia, Hefei, China, September 21-22, 2018, Proceedings, Part II   180 - 189   2018年

      詳細を見る

      担当区分:筆頭著者, 責任著者   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-00767-6_17

      researchmap

    • Fast super-resolution with iterative-guided back projection for 3D MR images 査読有り

      Yutaro Iwamoto, Xian-Hua Han, Akihiko Shiino, Yen-Wei Chen

      Progress in Biomedical Optics and Imaging - Proceedings of SPIE10574   105741T   2018年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE  

      Multimodal magnetic resonance images (e.g., T1-weighted image (TIWI) and T2-weighted image (T2WI)) are used for accurate medical imaging analysis. Different modal images have different resolution depending on pulse sequence parameters under limited data acquisition time. Therefore, interpolation methods are used to match the low-resolution (LR) image with the high-resolution (HR) image. However, the interpolation causes blurring that affects analysis accuracy. Although some recent works such as non-local-means (NLM) filter have manifested impressive super-resolution (SR) performance with available HR modal images, the filter has high computational cost. Therefore, we propose a fast SR framework with iterative-guided back projection, which incorporates iterative back projection with a guided filter (GF) method for resolution enhancement of LR images (e.g., T2WI) by referring HR images in another modality image (e.g., T1WI). The proposed method not only achieves both high accuracy than conventional interpolation methods and original GF and computational efficiency by applying an integral 3D image technique. In addition, although the proposed method is slightly inferior in accuracy visually than the state-of-the-art NLM filter, it can run 22 times faster than the state-of-theart method in expanding three times in the slice-select direction from 180 × 216 × 60 voxels to 180 × 216 × 180 voxels. The computational time of our method is about 1 min only. Therefore, the proposed method will be applied to various applications in practice, including not only multimodal MR images but also multimodal image analysis such as computed tomography (CT) and positron emission tomography (PET).

      DOI: 10.1117/12.2285336

      Scopus

      researchmap

    • Self-taught Learning with Residual Sparse Autoencoders for HEp-2 Cell Staining Pattern Recognition. 査読有り 国際共著

      Xian-Hua Han, JiandDe Sun, Lanfen Lin, Yen-Wei Chen

      Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16, 2018, Proceedings   134 - 142   2018年

      詳細を見る

      担当区分:筆頭著者, 責任著者   出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-00919-9_16

      researchmap

    • Multi-scale Residual Network with Two Channels of Raw CT Image and Its Differential Excitation Component for Emphysema Classification. 査読有り 国際共著

      Liying Peng, Lanfen Lin, Hongjie Hu, Huali Li, Qingqing Chen, Dan Wang, Xian-Hua Han, Yutaro Iwamoto, Yen-Wei Chen

      Deep Learning in Medical Image Analysis - and - Multimodal Learning for Clinical Decision Support - 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 20, 20   38 - 46   2018年

      詳細を見る

      出版者・発行元:Springer  

      DOI: 10.1007/978-3-030-00889-5_5

      researchmap

    • SSF-CNN: Spatial and Spectral Fusion with CNN for Hyperspectral Image Super-Resolution. 査読有り

      Xian-Hua Han, Boxin Shi, Yinqiang Zheng

      2018 IEEE International Conference on Image Processing, ICIP 2018, Athens, Greece, October 7-10, 2018   2506 - 2510   2018年

      詳細を見る

      担当区分:筆頭著者, 責任著者   出版者・発行元:IEEE  

      DOI: 10.1109/ICIP.2018.8451142

      researchmap

    • Tensor Sparse Representation of Temporal Features for Content-Based Retrieval of Focal Liver Lesions Using Multi-phase Medical Images 査読有り 国際共著

      Jian Wang, Xian-Hua Han, Yingying Xu, Lanfen Lin, Hongjie Hu, Chongwu Jin, Yen-Wei Chen

      Proceedings - 2017 IEEE International Symposium on Multimedia, ISM 20172017-   507 - 510   2017年12月28日

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Institute of Electrical and Electronics Engineers Inc.  

      Content Based Image Retrieval (CBIR) systems that search similar images in a large database are attracting more and more research interests recently, and have been applied to medical image characterization for expert's experience sharing. One challenging task in CBIR is how to extract features for effective image representation. Therein sparse coding technique has been proven to be an effective way to learn inherent structure features for image analysis. However, it is necessary to first vectorize the 2- or 3-dimensional spatial structure for analysis with sparse coding, and then destroy the spatial relation of nearby voxels. In this study, we propose a multilinear sparse coding method to learn features from multi-dimensional medical images. We regard high dimensional local structures as tensors and propose a K-CP (CANDECOMP/PARAFAC) algorithm to learn a tensor dictionary in an iterative way. With the learned tensor dictionary, sparse coefficients of tensor local structures are calculated by multilinear orthogonal matching pursuit (MOMP) algorithm, which is an extended multilinear version of the conventional linear OMP. The proposed multilinear sparse coding method is prospected to be more efficient and effective for inherent feature extraction compared with conventional linear methods. The proposed method is applied to a CBIR system for retrieval of focal liver lesions (FLLs) using a medical database consisting of contrast-enhanced multi-phase computer-tomography (CT) images. Experiments show that the constructed CBIR with multilinear sparse coding method can achieve promising retrieval performance.

      DOI: 10.1109/ISM.2017.100

      Scopus

      researchmap

    • Hyper-spectral image super-resolution using non-negative spectral representation with data-guided sparsity 査読有り

      Xian-Hua Han, Jan Wang, Boxin Shi, Yinqiang Zheng, Yen-Wei Chen

      Proceedings - 2017 IEEE International Symposium on Multimedia, ISM 20172017-   500 - 506   2017年12月28日

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Institute of Electrical and Electronics Engineers Inc.  

      Hyperspectral imaging has great potential for understanding the characteristics of different materials in many applications ranging from remote sensing to medical imaging. However, due to various hardware limitations, only low-resolution hyperspectral and high-resolution multi-spectral images can be available using existing imaging techniques. This study aims to generate a high-resolution hyperspectral image via fusion of the available LR-HS and HR-MS images. We propose a novel hyperspectral image superresolution method via non-negative sparse representation of reflectance spectral with adaptive sparsity constraint. By analyzing local content similarity of a focused pixel in the available high-resolution multi-spectral image, which can measure pixel material purity according to surrounding pixels, we generate a sparsity map for guiding non-negative sparse coding optimization procedure of the spectral representation called non-negative spectral representation with data-guided sparsity. Since the proposed method adaptively adjust the sparsity in the spectral representation based on the local content of the available high-resolution multi-spectral image, it can produce more robust spectral representation for recovering the target high-resolution hyper-spectral image. Comprehensive experiments on two public hyperspectral datasets validate that the proposed method achieves promising performances compared with the existing state of the art methods.

      DOI: 10.1109/ISM.2017.99

      Scopus

      researchmap

    • Generalized Aggregation of Sparse Coded Multi-Spectra for Satellite Scene Classification 査読有り 国際誌

      Xian-Hua Han, Yen-Wei Chen

      ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION6 ( 6 )   2017年6月

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:MDPI AG  

      Satellite scene classification is challenging because of the high variability inherent in satellite data. Although rapid progress in remote sensing techniques has been witnessed in recent years, the resolution of the available satellite images remains limited compared with the general images acquired using a common camera. On the other hand, a satellite image usually has a greater number of spectral bands than a general image, thereby permitting the multi-spectral analysis of different land materials and promoting low-resolution satellite scene recognition. This study advocates multi-spectral analysis and explores the middle-level statistics of spectral information for satellite scene representation instead of using spatial analysis. This approach is widely utilized in general image and natural scene classification and achieved promising recognition performance for different applications. The proposed multi-spectral analysis firstly learns the multi-spectral prototypes (codebook) for representing any pixel-wise spectral data, and then, based on the learned codebook, a sparse coded spectral vector can be obtained with machine learning techniques. Furthermore, in order to combine the set of coded spectral vectors in a satellite scene image, we propose a hybrid aggregation (pooling) approach, instead of conventional averaging and max pooling, which includes the benefits of the two existing methods, but avoids extremely noisy coded values. Experiments on three satellite datasets validated that the performance of our proposed approach is very impressive compared with the state-of-the-art methods for satellite scene classification.

      DOI: 10.3390/ijgi6060175

      researchmap

    • HEp-2 staining pattern recognition using stacked fisher network for encoding weber local descriptor 査読有り 国際誌

      Xian-Hua Han, Yen-Wei Chen

      PATTERN RECOGNITION63   542 - 550   2017年3月

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:ELSEVIER SCI LTD  

      This study addresses the recognition problem of the HEp-2 cell using indirect immunofluorescent (IIF) image analysis, which can indicate the presence of autoimmune diseases by finding antibodies in the patient serum. Generally, the method used for IIF analysis remains subjective, and depends too heavily on the experience and expertise of the physician. This study aims to explore an automatic HEp-2 cell recognition system, in which how to extract highly discriminate visual features plays a key role in this recognition application. In order to realize this purpose, our main efforts include: (1) a simple but robust local descriptor without any quantization for local patch representation; (2)A transformation of the difference values between the surrounding pixels and the center one to the perception degree, which is based on the fact that human perception for disguising a pattern depends not only on the absolute intensity of the stimulus but also on the relative variance of the stimulus; called as Weber local descriptor (WLD); (3) a data-driven coding strategy with a parametric probability process, and the extraction of not only low- but also high-order statistics for image representation called as Fisher vector; (4) the stacking of the Fisher network into multi-layer framework for more discriminate feature. Experiments using the open HEp-2 cell dataset released in the ICIP2013 contest validate that the proposed strategy can achieve a much better performance than the state-of-the-art approaches, and that the achieved recognition error rate is even very significantly below the observed intra-laboratory variability.

      DOI: 10.1016/j.patcog.2016.09.025

      researchmap

    • Phenotype Analysis Method for Identification of Gene Functions Involved in Asymmetric Division of Caenorhabditis elegans 査読有り 国際誌

      SiHai Yang, Xian-Hua Han, Tohsato Y, Kyoda K, Onami S, Nishikawa I, Chen Y

      J Comput Biol.24 ( 5 ) 436 - 446   2017年2月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)  

      DOI: 10.1089/cmb.2016.0210

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/jcb/jcb24.html#YangHTKONC17

    • Robust hepatic vessel segmentation using multi deep convolution network 査読有り 国際共著

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Amir Hossein Foruzan, Lanfen Lin, Yen-Wei Chen

      Progress in Biomedical Optics and Imaging - Proceedings of SPIE10137   2017年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE  

      Extraction of blood vessels of the organ is a challenging task in the area of medical image processing. It is really difficult to get accurate vessel segmentation results even with manually labeling by human being. The difficulty of vessels segmentation is the complicated structure of blood vessels and its large variations that make them hard to recognize. In this paper, we present deep artificial neural network architecture to automatically segment the hepatic vessels from computed tomography (CT) image. We proposed novel deep neural network (DNN) architecture for vessel segmentation from a medical CT volume, which consists of three deep convolution neural networks to extract features from difference planes of CT data. The three networks have share features at the first convolution layer but will separately learn their own features in the second layer. All three networks will join again at the top layer. To validate effectiveness and efficiency of our proposed method, we conduct experiments on 12 CT volumes which training data are randomly generate from 5 CT volumes and 7 using for test. Our network can yield an average dice coefficient 0.830, while 3D deep convolution neural network can yield around 0.7 and multi-scale can yield only 0.6.

      DOI: 10.1117/12.2253811

      Scopus

      researchmap

    • An Improved Random Walker with Bayes Model for Volumetric Medical Image Segmentation 査読有り 国際共著 国際誌

      Chunhua Dong, Xiangyan Zeng, Lanfen Lin, Hongjie Hu, Xianhua Han, Masoud Naghedolfeizi, Dawit Aberra, Yen-Wei Chen

      JOURNAL OF HEALTHCARE ENGINEERING   2017年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:HINDAWI LTD  

      Random walk (RW) method has been widely used to segment the organ in the volumetric medical image. However, it leads to a very large-scale graph due to a number of nodes equal to a voxel number and inaccurate segmentation because of the unavailability of appropriate initial seed point setting. In addition, the classical RW algorithm was designed for a user to mark a few pixels with an arbitrary number of labels, regardless of the intensity and shape information of the organ. Hence, we propose a prior knowledge-based Bayes random walk framework to segment the volumetric medical image in a slice-by-slice manner. Our strategy is to employ the previous segmented slice to obtain the shape and intensity knowledge of the target organ for the adjacent slice. According to the prior knowledge, the object/background seed points can be dynamically updated for the adjacent slice by combining the narrow band threshold (NBT) method and the organ model with a Gaussian process. Finally, a high-quality image segmentation result can be automatically achieved using Bayes RW algorithm. Comparing our method with conventional RW and state-of-the-art interactive segmentation methods, our results show an improvement in the accuracy for liver segmentation (p < 0 001).

      DOI: 10.1155/2017/6506049

      researchmap

    • Robust Hepatic Vessel Segmentation using Multi Deep Convolution Network 査読有り 国際共著

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Amir Hossein Foruzan, Lanfen Lino, Yen-Wei Chen

      MEDICAL IMAGING 2017: BIOMEDICAL APPLICATIONS IN MOLECULAR, STRUCTURAL, AND FUNCTIONAL IMAGING10137   2017年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      Extraction of blood vessels of the organ is a challenging task in the area of medical image processing. It is really difficult to get accurate vessel segmentation results even with manually labeling by human being. The difficulty of vessels segmentation is the complicated structure of blood vessels and its large variations that make them hard to recognize. In this paper, we present deep artificial neural network architecture to automatically segment the hepatic vessels from computed tomography (CT) image. We proposed novel deep neural network (DNN) architecture for vessel segmentation from a medical CT volume, which consists of three deep convolution neural networks to extract features from difference planes of CT data. The three networks have share features at the first convolution layer but will separately learn their own features in the second layer. All three networks will join again at the top layer. To validate effectiveness and efficiency of our proposed method, we conduct experiments on 12 CT volumes which training data are randomly generate from 5 CT volumes and 7 using for test. Our network can yield an average dice coefficient 0.830, while 3D deep convolution neural network can yield around 0.7 and multi-scale can yield only 0.6.

      DOI: 10.1117/12.2253811

      researchmap

    • Sparse Codebook Model of Local Structures for Retrieval of Focal Liver Lesions Using Multiphase Medical Images 査読有り 国際共著 国際誌

      Jian Wang, Xian-Hua Han, Yingying Xu, Lanfen Lin, Hongjie Hu, Chongwu Jin, Yen-Wei Chen

      International Journal of Biomedical Imaging2017   2017年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:Hindawi Publishing Corporation  

      Characterization and individual trait analysis of the focal liver lesions (FLL) is a challenging task in medical image processing and clinical site. The character analysis of a unconfirmed FLL case would be expected to benefit greatly from the accumulated FLL cases with experts' analysis, which can be achieved by content-based medical image retrieval (CBMIR). CBMIR mainly includes discriminated feature extraction and similarity calculation procedures. Bag-of-Visual-Words (BoVW) (codebook-based model) has been proven to be effective for different classification and retrieval tasks. This study investigates an improved codebook model for the fined-grained medical image representation with the following three advantages: (1) instead of SIFT, we exploit the local patch (structure) as the local descriptor, which can retain all detailed information and is more suitable for the fine-grained medical image applications
      (2) in order to more accurately approximate any local descriptor in coding procedure, the sparse coding method, instead of K-means algorithm, is employed for codebook learning and coded vector calculation
      (3) we evaluate retrieval performance of focal liver lesions (FLL) using multiphase computed tomography (CT) scans, in which the proposed codebook model is separately learned for each phase. The effectiveness of the proposed method is confirmed by our experiments on FLL retrieval.

      DOI: 10.1155/2017/1413297

      Scopus

      researchmap

    • Bayesian saliency model for focal liver lesion enhancement and detection 査読有り

      Xian-Hua Han, Jian Wang, Yuu Konno, Yen-Wei Chen

      Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)10118   32 - 45   2017年

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer Verlag  

      Focal liver lesion enhancement and detection has an essential role for the computer-aided diagnosis and characterization of lesion regions in CT volume data. This paper proposes a novel focal lesion enhancement strategy by extracting a lesion saliency map, which represents the deviation degree of the uncommon or lesion tissue from the common tissues (liver and vessel) in CT volumes. The saliency map can be constructed by exploring the existing probability of lesion for any voxel. However, due to the large diversity of liver lesions, it is difficult to construct an universal model for all types of lesions. Therefore, this study proposes to construct probability models of the common tissues, which have comparably small variability even for different samples and is relatively easy to obtain the prototype regions even from the understudying CT volume. In order to robustly and flexibly characterize the common tissues, we explore a Bayesian framework by combining a general model, which is constructed oriented to all CT samples, and an adaptive model, which is constructed specific to the under-studying CT sample, for calculating the existing probability of the common tissues (liver or vessel). Then, the saliency map (the existing probability) of focal lesion can be deduced from that of liver or vessel. The advantages of our proposed strategy mainly include three aspects: (1) it only needs to prepare the prototypes of common tissue such as liver or vessel region, which are easily obtained in any CT liver volume
      (2) it proposes to combine the general and adaptive model as Bayesian framework for more robust and flexible characterization of the common tissue
      (3) dispensable to remove the other different structure such as vessel in liver volume as a pre-processing step. Experiments validate that the proposed Bayesian-based saliency model for focal liver lesion enhancement can perform much better than the conventional approaches such as EM, EM/MPM based lesion detection and segmentation methods.

      DOI: 10.1007/978-3-319-54526-4_3

      Scopus

      researchmap

    • Integration of spatial and orientation contexts in local ternary patterns for HEp-2 cell classification 査読有り 国際誌

      Xian-Hua Han, Yen-Wei Chen, Gang Xu

      PATTERN RECOGNITION LETTERS82   23 - 27   2016年10月

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:ELSEVIER SCIENCE BV  

      In this paper, we describe a novel image representation strategy for classifying HEp-2 cell patterns of fluorescence staining. Our proposed strategy extends local binary patterns (LBPs), which are state-of-the-art texture features, into local ternary patterns (LTPs) with data-driven thresholds according to Weber's law, a human perception principle; further, our approach incorporates the contexts of spatial and orientation co-occurrences among adjacent Weber-based local ternary patterns (WLTPs) for texture representation. The explored WLTP is formulated by adaptively quantizing differential values between neighborhood pixels and the focused pixel as negative or positive stimuli if the normalized differential values are large; otherwise the stimulus is set to 0. Our approach here is based on the fact that human perception of a distinguished pattern depends not only on the absolute intensity of the stimulus but also on the relative variance of the stimulus. By integrating spatial and orientation context information, we further propose a rotation invariant co-occurrence WLTP (RICWLTP) approach to be more discriminant for image representation. Through experiment on the open HEp-2 cell dataset used at the ICPR2014 contest, we confirmed that our proposed strategy can greatly improve recognition performance or achieve comparable performance as compared with state-of-the-art LBP-based descriptor, the conventional LTP, and adaptively codebook/model based methods. (C) 2016 Elsevier B.V. All rights reserved.

      DOI: 10.1016/j.patrec.2016.02.004

      J-GLOBAL

      researchmap

    • Dual-band polarization angle independent 90 degrees polarization rotator using chiral metamaterial 査読有り 国際共著

      Hailin Cao, Huan Chen, Xiaodong Wu, Yuwei Pi, Junjie Liu, Hang Xu, Xiaoheng Tan, Jianmei Lei, Xian-Hua Han

      IEICE ELECTRONICS EXPRESS13 ( 15 )   2016年8月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      The polarization angle-independent, dual-band, 90 degrees polarization rotator presented in this paper consisted of a bilayered chiral metamaterial composed of twisted capacity-loaded I-shaped electric field-coupled resonators in C4 symmetry. Simulated and measured results consistently demonstrated that the rotator exhibited extremely low loss and high polarization ratio under dual-band conditions. We also systematically investigated the dependence of the electromagnetic response of the loaded structure on the geometric parameters. The proposed simple, easily fabricated model of a chiral metamaterial may be used in further polarization rotator applications.

      DOI: 10.1587/elex.13.20160583

      researchmap

    • Nuclear detection in 4D microscope images of a developing embryo using an enhanced probability map of top-ranked intensity-ordered descriptors.

      Xian-Hua Han, Yukako Tohsato, Koji Kyoda, Shuichi Onami, Ikuko Nishikawa, Yen-Wei Chen

      IPSJ Trans. Comput. Vis. Appl.8   8 - 8   2016年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1186/s41074-016-0010-3

      researchmap

      その他リンク: https://dblp.uni-trier.de/db/journals/ipsjtcva/ipsjtcva8.html#HanTKONC16

    • A framework for probabilistic atlas-based organ segmentation 査読有り 国際共著

      Chunhua Dong, Yen-Wei Chen, Amir Hossein Foruzan, Xian-Hua Han, Tomoko Tateyama, Xing Wu

      Progress in Biomedical Optics and Imaging - Proceedings of SPIE9784   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE  

      Probabilistic atlas based on human anatomical structure has been widely used for organ segmentation. The challenge is how to register the probabilistic atlas to the patient volume. Additionally, there is the disadvantage that the conventional probabilistic atlas may cause a bias toward the specific patient study due to a single reference. Hence, we propose a template matching framework based on an iterative probabilistic atlas for organ segmentation. Firstly, we find a bounding box for the organ based on human anatomical localization. Then, the probabilistic atlas is used as a template to find the organ in this bounding box by using template matching technology. Comparing our method with conventional and recently developed atlas-based methods, our results show an improvement in the segmentation accuracy for multiple organs (p &lt
      0:00001).

      DOI: 10.1117/12.2217340

      Scopus

      researchmap

    • Coccurrence Statistics of Local Ternary Patterns for HEp-2 Cell Classification 査読有り

      Xian-Hua Han, Yen-Wei Chen, Gang Xu

      INNOVATION IN MEDICINE AND HEALTHCARE 201545   205 - 213   2016年

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      In this paper, we describe a novel image representation strategy for classifying HEp-2 cell patterns of fluorescence staining. Our proposed strategy extends local binary patterns (LBPs), which are state-of-the-art texture features, into local ternary patterns (LTPs) with data-driven thresholds according to Weber's law, a human perception principle; further, our approach incorporates the contexts of spatial and orientation co-occurrences among adjacent Weber-based local ternary patterns (WLTPs) for texture representation. The explored WLTP is formulated by adaptively quantizing differential values between neighborhood pixels and the focused pixel as negative or positive stimuli if the normalized differential values are large; otherwise the stimulus is set to 0. Our approach here is based on the fact that human perception of a distinguished pattern depends not only on the absolute intensity of the stimulus but also on the relative variance of the stimulus. By integrating spatial and orientation context information, we further propose a rotation invariant co-occurrence WLTP (RICWLTP) approach to be more discriminant for image representation. Through experiment on the open HEp-2 cell dataset used at the ICIP2013 contest, we confirmed that our proposed strategy can greatly improve recognition performance or achieve comparable performance as compared with state-of-the-art LBP-based descriptor, the conventional LTP, and adaptively codebook/model based methods.

      DOI: 10.1007/978-3-319-23024-5_19

      researchmap

    • Automatic Registration of Deformable Organs in Medical Volume Data by Exhaustive Search 査読有り

      Masahiro Isobe, Shota Niga, Kei Ito, Xian-Hua Han, Yen-Wei Chen, Gang Xu

      INNOVATION IN MEDICINE AND HEALTHCARE 201545   309 - 320   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      This paper proposes a novel framework for fully automatic localization of deformable organs in medical volume data, which can obtain not only the position but also simultaneously the orientation and deformation of the organ to be searched, without the need to segment the organ first. The problem is defined as one of minimizing the sum of squared distances between the organ model's surface points and their closest surface points extracted from the input volume data. The geometric alignment, or so-called registration, of three-dimensional models by least square minimization always has the problem of initial states. We argue that the only way to solve this problem is by the exhaustive search. However, the exhaustive search takes much computational cost. In order to reduce the computational cost, we make efforts in the following three ways: (1) a uniform sampling over 3D rotation group; (2) Pyramidal search for all parameters; (3) Construction of a distance function for efficiently finding closest points. We have finished experiments for searching the six parameters for position and orientation, and the results show that the proposed framework can achieve correct localization of organs in the input data even with very large amounts of noise. We are currently expanding the system to localize organs with large deformation by adding and searching parameters representing scaling and deformation.

      DOI: 10.1007/978-3-319-23024-5_28

      researchmap

    • Bayesian Model for Liver Tumor Enhancement 査読有り 国際共著

      Yu Konno, Xian-Hua Han, Lanfen Lin, Hongjie Hu, Yitao Liu, Wenchao Zhu, Yen-Wei Chen

      INNOVATION IN MEDICINE AND HEALTHCARE 201660   227 - 235   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER INT PUBLISHING AG  

      Automatic liver lesion enhancement and detection has an essential role for the computer-aided diagnosis of liver tumor in CT volume data. This paper proposes a novel lesion enhancement strategy using Bayesian framework by combining the lesion probabilities based on an adaptive non-parametric model with the processed test volume and the constructed common non-lesion models with prepared liver database. Due to the large variation of different lesion tissues, it is difficult to obtain the common lesion prototypes from liver volumes, and thus this paper investigates a lesion-training-data free strategy by only constructing the healthy liver and vessel prototypes using local patches, which can be extracted from any slice of the test liver volume, and is also easy to prepare the common training non-lesion samples for all volumes. With the healthy liver and vessel prototypes from the test volume, an adaptive non-parametric model is constructed for estimating the lesion possibility, which is considered as the pixel likelihood to lesion region; the common model constructed using the pre-prepared liver database is used to estimate the pixel probability, which is defined as prior knowledge due to the used unvaried model. Finally, the posterior probabilities based on Bayesian theory are achieved for enhancing lesion regions. Experimental results validate that the proposed framework can not only detect almost small lesion regions but also greatly reduce falsely detect regions.

      DOI: 10.1007/978-3-319-39687-3_22

      researchmap

    • Combined Density, Texture and Shape Features of Multi-phase Contrast-Enhanced CT Images for CBIR of Focal Liver Lesions: A Preliminary Study 国際共著

      Yingying Xu, Lanfen Lin, Hongjie Hu, Huajun Yu, Chongwu Jin, Jian Wang, Xianhua Han, Yen-Wei Chen

      INNOVATION IN MEDICINE AND HEALTHCARE 201545   215 - 224   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      Recently, content-based image retrieval (CBIR) in medical applications has attracted a lot of attentions. In this paper, we present a preliminary study on CBIR of focal liver lesions based on combined density, texture and shape features of multi-phase contrast-enhanced CT volumes. We improve the existing method from following two aspects: (1) in order to improve the retriev al accuracy, we propose a novel 3D shape feature for CBIR of liver lesions in addition to conventional density and texture features; (2) in order to reduce the computation time, we propose an improved local binary pattern, which is called imLBP, as the 3D texture feature. The effectiveness of our proposed method has been validated with real clinical datasets.

      DOI: 10.1007/978-3-319-23024-5_20

      researchmap

    • A Novel and Fast Connected Component Count Algorithm Based on Graph Theory 査読有り 国際共著

      Sihai Yang, Duansheng Chen, Xianhua Han, Yenwei Chen

      2016 17TH IEEE/ACIS INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING, ARTIFICIAL INTELLIGENCE, NETWORKING AND PARALLEL/DISTRIBUTED COMPUTING (SNPD)   643 - 647   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      A fast component-counting algorithm is proposed based on graph theory in this paper. We derived a formulation to count faces in a plane given only the vertices based on Euler polyhedron formula. Vertices with degree no more than two are ineffective in counting components. With the derived formula, a graph component counting algorithm is constructed based only on searching cross points whose degree are no less than three. When applied to a two-dimensional binary image, the proposed method divides an image into patches of same size and decides which of them will be used in counting by searching the circumferential pixels of each patch. If the number of component edges within the circumferential pixels of a patch is no less than three, then the patch will be used in counting. After determining all the vertices with degree no less than three, the number of components can be calculated by the formula. Because only a small number of pixels are investigated in the process, the computational time is very fast. The disconnection of edges in an image is one of the main reasons which causes miscounts for scanning-based algorithms. The difficulty, however, can be naturally overcome by the proposed algorithm because disconnected points will be identified as futile pixels in the algorithm. Experimental results show the algorithm is more efficient than existing methods. When applied to images with disconnected edges, the counted number given by scanning-based algorithms is much smaller than the correct number whereas the proposed algorithm obtains satisfactory results.

      DOI: 10.1109/SNPD.2016.7515972

      researchmap

    • A framework for probabilistic atlas-based organ segmentation 査読有り 国際共著

      Chunhua Dong, Yen-Wei Chen, Amir Hossein Foruzan, Xian-Hua Han, Tomoko Tateyama, Xing Wu

      MEDICAL IMAGING 2016: IMAGE PROCESSING9784   97842X   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      Probabilistic atlas based on human anatomical structure has been widely used for organ segmentation. The challenge is how to register the probabilistic atlas to the patient volume. Additionally, there is the disadvantage that the conventional probabilistic atlas may cause a bias toward the specific patient study due to a single reference. Hence, we propose a template matching framework based on an iterative probabilistic atlas for organ segmentation. Firstly, we find a bounding box for the organ based on human anatomical localization. Then, the probabilistic atlas is used as a template to find the organ in this bounding box by using template matching technology. Comparing our method with conventional and recently developed atlas-based methods, our results show an improvement in the segmentation accuracy for multiple organs (p < 0.00001).

      DOI: 10.1117/12.2217340

      researchmap

    • HEp-2 Cell Classification Using K-Support Spatial Pooling in Deep CNNs 査読有り 国際共著

      Xian-Hua Han, Jianmei Lei, Yen-Wei Chen

      DEEP LEARNING AND DATA LABELING FOR MEDICAL APPLICATIONS10008   3 - 11   2016年

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER INT PUBLISHING AG  

      This study addresses the recognition problem of the HEp-2 cell using indirect immunofluorescent (IIF) image analysis, which can facilitate the diagnosis of many autoimmune diseases by finding antibodies in the patient serum. Recently, a lot of automatic HEp-2 cell classification strategies including both shallow and deep methods have been developed, wherein the deep Convolutional Neural Networks (CNNs) have been proven to achieve impressive performance. However, the deep CNNs in general requires a fixed size of image as the input. In order to conquer the limitation of the fixed size problem, a spatial pyramid pooling (SPP) strategy has been proposed in general object recognition and detection. The SPP-net usually exploit max pooling strategies for aggregating all activated status of a specific neuron in a predefined spatial region by only taking the maximum activation, which achieved superior performance compared with mean pooling strategy in the traditional state-of-the-art coding methods such as sparse coding, linear locality-constrained coding and so on. However, the max pooling strategy in SPP-net only retains the strongest activated pattern, and would completely ignore the frequency: an important signature for identifying different types of images, of the activated patterns. Therefore, this study explores a generalized spatial pooling strategy, called K-support spatial pooling, in deep CNNs by integrating not only the maximum activated magnitude but also the response magnitude of the relatively activated patterns of a specific neuron together. This proposed K-support spatial pooling strategy in deep CNNs combines the popularly applied mean and max pooling methods, and then avoid awfully emphasizing of the maximum activation but preferring a group of activations in a supported region. The deep CNNs with the proposed K-support spatial pooling is applied for HEp-2 cell classification, and achieve promising performance compared with the state-of-the-art approaches.

      DOI: 10.1007/978-3-319-46976-8_1

      researchmap

    • Bag of Temporal Co-occurrence Words for Retrieval of Focal Liver Lesions Using 3D Multiphase Contrast-Enhanced CT Images 査読有り 国際共著

      Yingying Xu, Lanfen Lin, Hongjie Hu, Dan Wang, Yitao Liu, Jian Wang, Xianhua Han, Yen-Wei Chen

      2016 23RD INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)   2282 - 2287   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      Computer-aided diagnosis (CAD) systems have been verified to have the potential to assist radiologists in clinical diagnosis to detect and characterize focal liver lesions (FLLs) based on single-or multiphase contrast-enhanced computed tomography (CT) images. Features extracted from multiphase contrast-enhanced CT images carry more important diagnostic information i.e. enhancement pattern and demonstrate much stronger discriminative ability compared to those of single-phase CT images. In this paper, we propose a new method for multiphase image feature generation called the bag of temporal co-occurrence words (BoTCoW). A temporal co-occurrence image connecting intensity from multiphase images is constructed. Then the bag of visual word (BoVW) model is employed on the temporal co-occurrence images to extract temporal features. The proposed method effectively captures temporal enhancement information and demonstrates the distribution of the evolution patterns. The effectiveness of this method is validated in a retrieval system using 132 FLLs with confirmed pathology type. The preliminary results show that the proposed BoTCoW method outperforms the previously proposed temporal features and multiphase features based on the BoVW model.

      DOI: 10.1109/ICPR.2016.7899976

      researchmap

    • Super-resolution of 3D MR images and its application to brain segmentation 査読有り

      Yutaro Iwamoto, Xian-Hua Han, Akihiko Shiino, Yen-Wei Chen

      2016 9TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING, BIOMEDICAL ENGINEERING AND INFORMATICS (CISP-BMEI 2016)   838 - 841   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper presents super resolution (SR) of 3D MR images which is effective for brain segmentation as practical application. Brain segmentation is one of important tasks to analyze brain morphometry. An accurate brain segmentation helps improve accuracy of post-processing. The segmentation is affected by the resolution of 3D MR images. In particular, the resolution in slice-select direction is much lower than the in-plane direction in some type of 3D MR images (e.g. T2-weighted images and proton density images). Therefore, we apply a learning-based SR method using sparse representation and self-similarity for generating high-resolution (HR) in slice-select direction to be same resolution in in-plane direction. With the visualized evaluation, it can be seen that the segmentation accuracy in SR results are improved by reducing partial volume effect. With the quantitative evaluation, one can confirm that the DICE value of SR results are higher than conventional interpolation method.

      DOI: 10.1109/CISP-BMEI.2016.7852827

      researchmap

    • SIFT-Based Multi-Frame Super Resolution for 250 Million Pixel Images 査読有り

      Katsuhisa Ogawa, Yuri Yamaguchi, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen

      2016 9TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING, BIOMEDICAL ENGINEERING AND INFORMATICS (CISP-BMEI 2016)   834 - 837   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In this paper, we propose a SIFT-based multi-frame super resolution for 250 million pixel images. In the proposed method, we first use the SIFT operator to detect key points in each frame. Then we use a closest matching method to find the correspondence among multi-frame images. The corresponding key points are used to register multi-frame images to a reference image, which is randomly selected from the multi-frame images. After registration, we combine the aligned multi-frame images to form a high-quality and high-resolution image. We applied the proposed method to enhance the quality of 250 million pixel images, which is obtained by the Canon's 250Mpixel CMOS-image-sensor.

      DOI: 10.1109/CISP-BMEI.2016.7852826

      researchmap

    • Food Recognition by Combined Bags of Color Features and Texture Features 査読有り

      Shota Sasano, Xian-Hua Han, Yen-Wei Chen

      2016 9TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING, BIOMEDICAL ENGINEERING AND INFORMATICS (CISP-BMEI 2016)   815 - 819   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Food is one of the most important issues for human health. In order to manage the healthy dietary life, we are going to develop an auto food-log record system, which is based on auto food recognition technologies. To realize this system, we aim to propose a discriminated food image representation that can perform effective identification of food images in this paper. The conventional image representation mainly includes color and texture distributions (histogram), which are the statistical information based on uniformly quantized color or texture levels. However, these conventional techniques using uniform quantization of the on-hand color and texture in the image lead much information loss for reliably constructing the image. Therefor, this study proposes to characterize the color and texture information by incorporating the strategy of patch-based bag of features model. This technique can adaptively learn the representative color or texture (prototypes) from the food images for food recognition, and it is possible to recover a more reliable image using the learned prototypes. The experiments using our proposed approaches show that the recognition rate can be greatly improved compared with the conventional method.

      DOI: 10.1109/CISP-BMEI.2016.7852822

      researchmap

    • A Retrieval System for 3D Multi-Phase Contrast-Enhanced CT Images of Focal Liver Lesions Based on Combined Bags of Visual Words and Texture Words 査読有り 国際共著

      Yingying Xu, Lanfen Lin, Hongjie Hu, Dan Wang, Yitao Liu, Jian Wang, Xianhua Han, Yen-Wei Chen

      2016 9TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING, BIOMEDICAL ENGINEERING AND INFORMATICS (CISP-BMEI 2016)   806 - 810   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Content-based image retrieval (CBIR) technique for digital image searching has been applied in medical images which are called content-based medical image retrieval (CBMIR). In this paper, we combine texture, density and shape features for CBMIR based on 3D multi-phase contrast enhanced CT images according to radiologists' clinical experience. And a database of 132 focal liver lesions (FLLs) with confirmed pathology type is constructed in this work. We implement bag of visual words (BoVW) model to extract texture features from FLLs based on 3D local binary pattern (LBP) and combine it with conventional intensity-based BoVW. Density and temporal density are designed based on clinical observation. Principle Component Analysis (PCA) is used to extract the sphericity of the lesions as shape features. Experiments performed on the FLLs database show the efficiency of the proposed system and features.

      DOI: 10.1109/CISP-BMEI.2016.7852820

      researchmap

    • Simultaneous Segmentation of Multiple Organs Using Random Walks. 査読有り 国際共著

      Chunhua Dong, Yen-Wei Chen, Lanfen Lin, Hongjie Hu, Chongwu Jin, Huajun Yu, Xian-Hua Han, Tomoko Tateyama

      JIP24 ( 2 ) 320 - 329   2016年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.2197/ipsjjip.24.320

      researchmap

    • A principal component analysis based method to automatically inspect wear of throw-away tips 査読有り 国際誌

      Ting Wang, Rui Xu, Xianhua Han, Yen-Wei Chen, Yoshitomo Ishizaki, Masaru Miyamoto, Tomohito Hattori

      JOURNAL OF INTELLIGENT & FUZZY SYSTEMS31 ( 2 ) 903 - 913   2016年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IOS PRESS  

      The automatic inspection of throw-away tips is very important for quality control in precision cutting. We proposed an image processing based method for automatic inspection of the processing wear of throw-away tips. After image denoising, the proposed method utilized image-patch based principal component analysis method to enhance the cutting worn region while suppress the background region. Then the enhanced worn region was automatically segmented by a simple thresholding method followed by post-processing. The area of the segmented worn region was used as a measure of cutting wear degree. We collected three datasets of time-series images that recorded the processing of throw-away tips on a product line. One dataset was used to choose optimal parameters of the proposed method, and the other two datasets were used for evaluate its performances. Experimental results showed that the proposed method was able to inspect the cutting wear with high accuracy. Additionally, it was also showed that the proposed method outperformed the conventional thresholding based method.

      DOI: 10.3233/JIFS-169020

      researchmap

    • Non-rigid image registration with anatomical structure constraint for assessing locoregional therapy of hepatocellular carcinoma 査読有り 国際誌

      Chunhua Dong, Yen-wei Chen, Toshihito Seki, Ryosuke Inoguchi, Chen-Lun Lin, Xian-hua Han

      COMPUTERIZED MEDICAL IMAGING AND GRAPHICS45   75 - 83   2015年10月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:PERGAMON-ELSEVIER SCIENCE LTD  

      Purpose: Assessing the treated region with locoregional therapy (LT) provides valuable information for predicting hepatocellular carcinoma (HCC) recurrence. The commonly used of assessment method is inefficient because it only compares two-dimensional CF images manually. In our previous work, we automatically aligned the two CT volumes to evaluate the therapeutic efficiency using registration algorithms. The non-rigid registration is applied to capture local deformation, however, it usually destroys internal structure. Taking these into consideration, this paper proposes a novel non-rigid registration approach for evaluating LT of HCC to maintain the image integrity.
      Method: In our registration algorithm, a global affine transformation combined with localized cubic B-spline is used to estimate the significant non-rigid motions of two livers. The proposed method extends a classical non-rigid registration based on mutual information (MI) that uses an anatomical structure term to constrain the local deformation. The energy function can be defined based on the total one associated with the anatomical structure and deformation information. Optimal transformation is obtained by finding the equilibrium state in which the total energy is minimized, indicating that the anatomical landmarks have found their correspondences. Thus, we can use the same transformation to automatically transform the ablative region to the optimal position.
      Results: Registration accuracy is evaluated using the clinical data. Improved results are obtained with respect to all criteria in our proposed method (MI-LC) than those in the MI-based non-rigid registration. The landmark distance error (LDE) of MI-LC is decreased by an average of 3.93 mm compared to the case of MI-based registration. Moreover, it could be found regardless of how many landmarks applied in our proposed method, a significant reduction in LDE values using registrations based on MI-LC compared with those based on MI is confirmed.
      Conclusion: Our proposed approach can guarantee the continuity, the accuracy and the smoothness of structures by constraining the anatomical features. The results clearly indicate that our method can retain the local deformation of the image. In addition, it assures the anatomical structure stability. (C) 2015 Elsevier Ltd. All rights reserved.

      DOI: 10.1016/j.compmedimag.2015.08.003

      researchmap

    • Supervoxels based graph cut for medical organ segmentation 査読有り 国際共著

      Titinunt Kitrungrotsakul, Yen-Wei Chen, Xian-Hua Han, Lanfen Lin

      IFAC-PapersOnLine28 ( 20 ) 70 - 75   2015年9月1日

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      Organ segmentation is one of the most fundamental and challenging tasks in computer aid diagnosis system. Researches successfully working on interactive segmentation for medical image include graph cuts and random walk. However, graph cut-based organ segmentation for 3D medical volume data requires an optimization procedure of cutting the object/background regions on a very largescale graph, which not only consumes large amount of memory and but also requires expensive computational cost. This paper conquers the drawbacks of graph cut-based organ segmentation via instead of voxel with supervoxel as nodes for constructing graph in interactive 3D organ segmentation that can greatly reduces node number and connected edges
      voxels of medical data with similar intensity magnitude and near spatial relation are grouped into the supervoxels and such supervoxels are used as pseudo nodes of graph for cutting object (organ) and background regions, named as supervoxel-based graph cut. To validate the effectiveness and efficiency of the proposed method, we conduct experiments on 10 medical data, which possibly have tumors inside organ, or have abnormal deformed organ shape. The experimental results show that the proposed method is much superior than conventional graph cutbased method in term of accuracy, computational time and memory usage.

      DOI: 10.1016/j.ifacol.2015.10.117

      Scopus

      researchmap

    • High-Order Statistics of Weber Local Descriptors for Image Representation 査読有り 国際誌

      Xian-Hua Han, Yen-Wei Chen, Gang Xu

      IEEE TRANSACTIONS ON CYBERNETICS45 ( 6 ) 1180 - 1193   2015年6月

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC  

      Highly discriminant visual features play a key role in different image classification applications. This study aims to realize a method for extracting highly-discriminant features from images by exploring a robust local descriptor inspired by Weber's law. The investigated local descriptor is based on the fact that human perception for distinguishing a pattern depends not only on the absolute intensity of the stimulus but also on the relative variance of the stimulus. Therefore, we firstly transform the original stimulus (the images in our study) into a differential excitation-domain according to Weber's law, and then explore a local patch, called micro-Texton, in the transformed domain as Weber local descriptor (WLD). Furthermore, we propose to employ a parametric probability process to model the Weber local descriptors, and extract the higher-order statistics to the model parameters for image representation. The proposed strategy can adaptively characterize the WLD space using generative probability model, and then learn the parameters for better fitting the training space, which would lead to more discriminant representation for images. In order to validate the efficiency of the proposed strategy, we apply three different image classification applications including texture, food images and HEp-2 cell pattern recognition, which validates that our proposed strategy has advantages over the state-of-the-art approaches.

      DOI: 10.1109/TCYB.2014.2346793

      J-GLOBAL

      researchmap

    • Statistical Shape Model of the Liver and Its Application to Computer-Aided Diagnosis of Liver Cirrhosis 査読有り

      Mei Uetani, Tomoko Tateyama, Shinya Kohara, Hidetoshi Tanaka, Xian-Hua Han, Shuzo Kanasaki, Akira Furukawa, Yen-Wei Chen

      ELECTRICAL ENGINEERING IN JAPAN190 ( 4 ) 37 - 45   2015年3月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:WILEY-BLACKWELL  

      In recent years, there has been increasing interest in statistical shape modeling of human anatomy. The statistical shape model can capture the morphological variations of human anatomy. Since liver cirrhosis will cause significant morphological changes, the authors propose a computer-aided diagnosis method for liver cirrhosis based on statistical shape models. In the proposed method, the authors first construct a statistical shape model of the liver using 50 clinical CT datasets (25 sets of normal data and 25 sets of abnormal data). The authors apply the marching cubes algorithm to convert the segmented liver volume to a triangulated mesh surface containing 1000 vertex points. The coordinates of these vertex points are used to represent the 3D liver shape as a shape vector. After normalization and identification of correspondences between all datasets, principal component analysis (PCA) is employed to find the principal variation modes of the shape vectors. Then the authors propose a mode selection method based on class variations between the normal class and abnormal class. The authors found that the top two modes of class variations are most effective for the classification of normal and abnormal livers. The classification rate of abnormal livers and normal livers by the use of a simple linear discriminant function were 84% and 80%, respectively. (C) 2014 by Wiley Periodicals.

      DOI: 10.1002/eej.22668

      researchmap

    • Independent Component Analysis-Based Effective Prediction of O-linked Glycosylation Sites in Protein by Support Vector Machine 査読有り 国際共著

      Chu-Zheng Wang, Hong-Yang Ren, Xian-Hua Han, Yen-Wei Chen

      PROCEEDINGS OF 2015 4TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND NETWORK TECHNOLOGY (ICCSNT 2015)   365 - 368   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Glycosylation Site Prediction (GSP) research has witnessed a growing interest in proteomics. The high ability to GSP is helpful for better understanding the function of protein, theoretically. In this research, our aim is to explore a new method for improving the performance of GSP of O-glycosylation sites. We propose to utilize Independent Component Analysis (ICA) for feature selection and dimension reduction, and then use Support Vector Machine (SVM) for glycosylation site classification, in which our method is applied for two kinds of datasets in glycosylated site and nonglycosylated site. Compared with using other subspace-based method and SVM method, experimental results show that our new approach is feasible and effective with higher prediction accuracy.

      researchmap

    • スパース表現と自己相似性を用いた三次元医用画像の超解像処理

      岩本裕太郎, 韓先花, 椎野顕彦, 陳延偉

      電子情報通信学会論文誌DJ98-D   1312-1324   2015年

      詳細を見る

    • Interactive segmentation and visualization system for medical images on mobile devices 査読有り 国際誌

      Titinunt Kitrungrotsakul, Chunhua Dong, Tomoko Tateyama, Xian-Hua Han, Yen-Wei Chen

      JOURNAL OF ADVANCED SIMULATION IN SCIENCE AND ENGINEERING2 ( 1 ) 96 - 107   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:JAPAN SOC SIMULATION TECHNOLOGY-JSST  

      Ubiquitous computing is an important technology in medicine that is predicted to support doctors anywhere and anytime. To help achieve it, this paper develops the Interactive Segmentation and Visualization System for Medical images on Mobile devices (ISVS_M-2), which originally was designed to work on workstations, but also works on a wide range of mobile devices via a mobile client-server platform. The developed ISVS_M-2 basically consists of three modules: asegmentation module that is implemented on a server; commutation modules on both the server and mobile device; and interactive and visualization modules on the mobile device that not only give visualization of internal information of an organ model, but also interactively refine organ segmentation according to user experience. With interaction via a computer graphic interface on the mobile device, and communication via the mobile client-server platform, ISVS_M-2 offers users a novel and efficient approach to computer-aided medicine.

      DOI: 10.15748/jasse.2.96

      researchmap

    • Segmentation of liver and spleen based on computational anatomy models. 査読有り 国際共著 国際誌

      Chunhua Dong, Yen-Wei Chen, Amir Hossein Foruzan, Lanfen Lin, Xian-Hua Han, Tomoko Tateyama, Xing Wu, Gang Xu, Huiyan Jiang

      Comp. in Bio. and Med.67   146 - 160   2015年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1016/j.compbiomed.2015.10.007

      researchmap

    • Co-occurrence Context of the data-driven Quantized Local Ternary Patterns for Visual Recognition 査読有り

      Xian-Hua Han, Yen-Wei Chen, Gang xu

      Proceedings 3rd IAPR Asian Conference on Pattern Recognition ACPR 20159 ( 4 ) 820 - 824   2015年

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In this paper, we describe a novel local descriptor of image texture representation for visual recognition. The image features based on micro-descriptors such as local binary patterns (LBP) and local ternary patterns (LTP) have been very successful in number of applications including face recognition and texture analysis. Instead of binary quantization in LBP, LTP thresholds the differential values between a focused pixel and its neighborhood pixels into three graylevel, which can be explained as the active status (i.e., positively activated, negatively activated and not activated) of the neighborhood pixels compared to the focused pixel. However, regardless to the magnitude of the focused pixel, the thresholding strategy remains fixed, which would violate the principle of human perception. Therefore, in this study, we design LTP with a data-driven threshold according to Weber's law, a human perception principle; further, our approach incorporates the contexts of spatial and orientation co-occurrences (i.e., co-occurrence context) among adjacent Weber-based local ternary patterns (WLTPs) for texture representation. In order to validate efficiency of our proposed strategy, we apply to three different visual recognition applications including two texture datasets and one food image dataset, and prove the promising performance can be achieved compared with the state-of-the-art approaches.

      researchmap

    • HEp-2 staining pattern recognition using stacked fisher network for encoding Weber local descriptor 査読有り

      Xian-Hua Han, Yen-Wei Chen, Gang Xu

      Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)9352   85 - 93   2015年

      詳細を見る

      担当区分:筆頭著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer Verlag  

      This study addresses the recognition problem of the HEp-2 cell using indirect immunofluorescent (IIF) image analysis, which can indicate the presence of autoimmune diseases by finding antibodies in the patient serum. Generally, the method used for IIF analysis remains subjective, and depends too heavily on the experience and expertise of the physician. This study aims to explore an automatic HEp-2 cell recognition system, in which how to extract highly discriminate visual features plays a key role in this recognition application. In order to realize this purpose, our main efforts include: (1) a transformed excitation domain instead of the raw image domain, which is based on the fact that human perception for disguising a pattern depends not only on the absolute intensity of the stimulus but also on the relative variance of the stimulus
      (2) a simple but robust micro-texton without any quantization in the excitation domain, called as Weber local descriptor (WLD)
      (3) a data-driven coding strategy with a parametric probability process, and the extraction of not only low- but also high-order statistics for image representation called as Fisher vector
      (4) the stacking of the Fisher network into deep learning framework for more discriminate feature. Experiments using the open HEp-2 cell dataset released in the ICIP2013 contest validate that the proposed strategy can achieve a much better performance than the state-of-the-art approaches, and that the achieved recognition error rate is even very significantly below the observed intra-laboratory variability.

      DOI: 10.1007/978-3-319-24888-2_11

      Scopus

      researchmap

    • LIVER SEGMENTATION USING SUPERPIXEL-BASED GRAPH CUTS AND RESTRICTED REGIONS OF SHAPE CONSTRAINS 査読有り

      Titinunt Kilrungrotsakul, Xian-Hua Han, Yen-Wei Chen

      2015 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)   3368 - 3371   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Liver segmentation is one of the most fundamental and challenging tasks in computer aided diagnosis (CAD) system for liver diseases. Graph cut algorithms have been successfully applied to medical image segmentation of different organs for 3D volume data, which not only leads to very large-scale graph due to the same node number as voxel number, but also completely ignore some available organ shape priors. Thus, a slice by slice liver segmentation method by combining shape constraints according to previously slice segmentation has been proposed based on graph cut. However, the constructed graph scale is still large, and the computation of distance map from all voxel to the segmented shape leads to high cost. In order to explore an efficient and effective slice by slice segmentation method for liver, this paper proposes to apply clustering algorithm to firstly group slice pixels into superpixels as nodes for constructing graph, which not only greatly reduce the graph scale but also significantly speed up the optimization procedure of the graph. Furthermore, we restrict the regions near organ boundary as shape constraints, which can further reduce computational time. To validate effectiveness and efficiency of our proposed method, we conduct experiments on 10 CT volumes, most of which have tumors inside liver, and abnormal deformed shape of liver. Our method can yield an average dice coefficient: 0.94, about 659.22 second in computation, and take only 1.5GB in memory usage.

      DOI: 10.1109/ICIP.2015.7351428

      researchmap

    • Two-Step Learning Based Super Resolution and Its Application to 3D Medical Volumes 査読有り

      Yuto Kondo, Xian-Hua Han, Yen-Wei Chen

      2015 IEEE 4TH GLOBAL CONFERENCE ON CONSUMER ELECTRONICS (GCCE)   326 - 327   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In medical diagnosis, high resolution (HR) images are indispensable for giving more correct decision. The super resolution technique, which can generate HR images from LR images based on machine learning, attracts hot attention recently. However, the conventional learning based SR generally cannot recover high frequency information. In this paper, we integrate a further learning step into the conventional method, and proposes a two-step learning based SR, which is prospected to recover most high frequency information lost in the available LR input. Furthermore, we also propose to use HR axial plane images of input volumes as HR training data to reconstruct HR coronal plane and sagittal plane images.

      DOI: 10.1109/GCCE.2015.7398738

      researchmap

    • Generic and Specific Impression Estimation of Clothing Fabric Images Based on Machine Learning 査読有り 国際共著

      Yen-Wei Chen, Dingye Chen, Xian-hua Han, Xinyin Huang

      2015 12th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)   1753 - 1757   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Consumers' psychological feeling or impression is an important factor for product design. The impression estimation becomes an important issue. In this paper, we propose generic and specific impression estimation methods based on machine learning for cloth fabric images. We use a semantic differential (SD) method to measure the user's impression such as bright, warm while they viewing a cloth fabric image. We also extract both global and local features of cloth fabric images such as color and texture using computer vision techniques. Then we use support vector regression to model the mapping functions between the generic impression (or specific impression) and image features. The learned mapping functions are used to estimate the generic or specific impression of cloth fabric images.

      DOI: 10.1109/FSKD.2015.7382212

      researchmap

    • Automatic Inspection of Throw-Away Tips Based on Principal Component Analysis 査読有り

      Ting Wang, Xianhua Han, Rui Xu, Yen-Wei Chen, Yoshitomo Ishizaki, Masaru Miyamoto, Tomohito Hattori

      2015 12th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)   1741 - 1746   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      The automatic inspection of throw-away tips is very important for quality control in precision cutting. We proposed an image processing based method for automatic inspection of the processing wear of throw-away tips. Our proposed method utilized principal component analysis as a preprocessing to reduce the noise and enhance the contrast of the cutting wear region with the background region. Then the PCA transformed wear region is automatically segmented by a simple thresholding method. The area of the segmented wear region is used as a measure of cutting wear degree.

      DOI: 10.1109/FSKD.2015.7382210

      researchmap

    • A Knowledge-based Interactive Liver Segmentation using Random Walks 査読有り 国際共著

      Chunhua Dong, Yen-Wei Chen, Tomoko Tateyama, Xian-hua Han, Lanfen Lin, Hongjie Hu, Chongwu Jin, Huajun Yu

      2015 12th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)   1731 - 1736   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      A random walks-based (RW) segmentation method has been gaining popularity in recent years with its ability to interactively segment the objects with minimal guidance. It has potential applications in segmenting the 3D image. However, due to the large computational burden of the classical RW algorithm, it is a challenge to use this algorithm to segment 3D medical images interactively. Hence, a knowledge-based segmentation framework for the liver is proposed based on random walks and narrow band threshold (RWNBT). Our strategy is to employ the previous segmented slice to achieve a prior knowledge (the shape and intensity constraints) of liver for automatic segmentation of the adjacent slice. With a small number of user-defined seeds, we can obtain the segmentation results of the start slice in the volume which would be used as the prior knowledge of the segmented organ. According to this intensity constraints, the "Candidate Pixels" image can be generated by thresholding the organ models with Gaussian Mixture Model (GMM), which can remove the noise and non-liver parts. Furthermore, the object/background seeds can be dynamically updated for the adjacent slice by combining a narrow band threshold (NBT) method and the shape constrains. Finally, a combinational random walker algorithm is applied to automatically segment the whole volume in a slice-by-slice manner. Comparing our method with conventional RW and the state-of-the-art interactive segmentation methods, our results show an improvement in the accuracy for liver segmentation.

      DOI: 10.1109/FSKD.2015.7382208

      researchmap

    • A Robust Registration Method using Huber ICP and Low Rank and Sparse Decomposition 査読有り

      Qiaochu Zhao, Xian-Hua Han, Yen-Wei Chen

      2015 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA)   744 - 752   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper proposes a robust registration and alignment framework for multiple point clouds using low rank and sparse decomposition. A coarse registration phase utilizing Huber-ICP is firstly performed to roughly align all the point clouds to a same location, and then sparse and low rank decomposition is applied to extract the low rank subspace of all the point clouds, which is expected to be outlier and loss data free. Finally, a fine registration procedure can be carried out between each point clouds from this low rank space to not only a more accurate registration result but also a more precise correspondence. Robustness of our method for outliers contained in point clouds is verified through manufactured data and it also shows that an effective result can still be achieved even when some points in the cloud are lost.

      DOI: 10.1109/APSIPA.2015.7415371

      researchmap

    • Co-occurrence Context of the data-driven Quantized Local Ternary Patterns for Visual Recognition 査読有り

      Xian-Hua Han, Yen-Wei Chen, Gang xu

      Proceedings 3rd IAPR Asian Conference on Pattern Recognition ACPR 2015   820 - 824   2015年

      詳細を見る

      担当区分:筆頭著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In this paper, we describe a novel local descriptor of image texture representation for visual recognition. The image features based on micro-descriptors such as local binary patterns (LBP) and local ternary patterns (LTP) have been very successful in number of applications including face recognition and texture analysis. Instead of binary quantization in LBP, LTP thresholds the differential values between a focused pixel and its neighborhood pixels into three graylevel, which can be explained as the active status (i.e., positively activated, negatively activated and not activated) of the neighborhood pixels compared to the focused pixel. However, regardless to the magnitude of the focused pixel, the thresholding strategy remains fixed, which would violate the principle of human perception. Therefore, in this study, we design LTP with a data-driven threshold according to Weber's law, a human perception principle; further, our approach incorporates the contexts of spatial and orientation co-occurrences (i.e., co-occurrence context) among adjacent Weber-based local ternary patterns (WLTPs) for texture representation. In order to validate efficiency of our proposed strategy, we apply to three different visual recognition applications including two texture datasets and one food image dataset, and prove the promising performance can be achieved compared with the state-of-the-art approaches.

      DOI: 10.1109/ACPR.2015.7486617

      researchmap

    • Nuclear Detection in 4D Microscope Images Using Enhanced Probability Map of Top-ranked Intensity-ordered Descriptors 査読有り

      Xian-Hua Han, Yukako Tohsato, Koji Kyoda, Shuichi Onami, Ikuko Nishikawa, Yen-Wei Chen

      Proceedings 3rd IAPR Asian Conference on Pattern Recognition ACPR 20158 ( 1 ) 554 - 558   2015年

      詳細を見る

      担当区分:筆頭著者, 責任著者   記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Nuclear detection in embryos is an indispensable process for quantitative analysis of the development of multicellular organisms. Due to the overlap in the distribution of nuclear and cytoplasmic intensities and the large variation even within the same type of tissues of different embryos, it is difficult to separate nuclear regions from the surrounding cytoplasmic region in differential interference contrast (DIC) microscope image. This study explores a discriminative representation of texton around a fixed pixel, called Top-ranked Intensity-ordered Descriptor (TRIOD), which is prospected to distinguish the smoothed texture in nucleus from the irregular texture in cytoplasm containing yolk granules. Then, a probability process is employed to model nuclear TRIOD prototypes, and the enhanced nuclear probability map can be constructed with the TRIODs of all pixels in a DIC microscope image. Finally, distance regularized level set method is applied to refine the initial localization by simply thresholding on the enhanced probability map. Experimental results show that the proposed strategy can give much better performance for segmentation of nuclear regions than the conventional strategies.

      DOI: 10.1109/ACPR.2015.7486564

      Scopus

      researchmap

    • Discriminant Statistical Analysis of Local Facial Geometrical Regions 査読有り

      Misae Nakatsu, Xian-Hua Han, Ryosuke Kimura, Yen-Wei Chen

      Proceedings 3rd IAPR Asian Conference on Pattern Recognition ACPR 2015   351 - 355   2015年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      The residences of Japanese Archipelago mainly include the three human populations; the Ainu, the Mainland Japanese and the Ryukyuan, which can he inferred by genome-wide single-nucleotide polymorphism (SNP) data and characterized by generic base sequences. In the other hand, the genetic association of human :facial morphological variation recently becomes a more and more active research field, aims to quantitatively analyze the possible relation measure between the gene base and a kind of facial morphological variations. This study attempts to explore the discriminated phenotype features of the common facial morphological variations between the Mainland Japanese and the Ryukyuan; the difference of phenotype features between these two populations is prospected to infer different gene base sequences. In order to explore the facial phenotype features, we propose a framework of local statistical analysis for adjacent geometrical regions of 3D facial images. Therein, we firstly select the surface points with higher distinguishable values based fisher linear discriminate analysis as discriminated coordinate vectors, and further cluster them into local geometrical groups for morphological analysis. The extracted local phenotype features are applied for identification of two populations, and achieve the comparable or better performances than the global phenotype feature, which manifests the possibility fOr association analysis between local morphological phenotype and the genes.

      DOI: 10.1109/ACPR.2015.7486524

      researchmap

    • Sparse and Low-Rank Matrix Decomposition for Local Morphological Analysis to Diagnose Cirrhosis 査読有り

      Junping Deng, Xian-Hua Han, Yen-Wei Chen, Gang Xu, Yoshinobu Sato, Masatoshi Hori, Noriyuki Tomiyama

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE97D ( 12 ) 3210 - 3221   2014年12月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      Chronic liver disease is a major worldwide health problem. Diagnosis and staging of chronic liver diseases is an important issue. In this paper, we propose a quantitative method of analyzing local morphological changes for accurate and practical computer-aided diagnosis of cirrhosis. Our method is based on sparse and low-rank matrix decomposition, since the matrix of the liver shapes can be decomposed into two parts: a low-rank matrix, which can be considered similar to that of a normal liver, and a sparse error term that represents the local deformation. Compared with the previous global morphological analysis strategy based on the statistical shape model (SSM), our proposed method improves the accuracy of both normal and abnormal classifications. We also propose using the norm of the sparse error term as a simple measure for classification as normal or abnormal. The experimental results of the proposed method are better than those of the state-of-the-art SSM-based methods.

      DOI: 10.1587/transinf.2014EDP7180

      researchmap

    • High-Order Statistics of Microtexton for HEp-2 Staining Pattern Classification 査読有り

      Xian-Hua Han, Jian Wang, Gang Xu, Yen-Wei Chen

      IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING61 ( 8 ) 2223 - 2234   2014年8月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC  

      This study addresses the classification problem of the HEp-2 cell using indirect immunofluorescent (IIF) image analysis, which can indicate the presence of autoimmune diseases by finding antibodies in the patient serum. Generally, the method used for IIF analysis remains subjective, and depends too heavily on the experience and expertise of the physician. Recently, studies have shown that it is possible to identify the cell patterns using IIF image analysis and machine learning techniques. However, it still has large gap in recognition rates to the physical experts' one. This paper explores an approach in which the discriminative features of HEp-2 cell images in IIF are extracted and then, the patterns of the HEp-2 cell are identified using machine learning techniques. Motivated by the progress in the research field of computer vision, as a result of which small local pixel pattern distributions can now be highly discriminative, the proposed strategy employs a parametric probability process to model local image patches (textons: microstructures in the cell image) and extract the higher-order statistics of the model parameters for the image description. The proposed strategy can adaptively characterize the microtexton space of HEp-2 cell images as a generative probability model, and discover the parameters that yield a better fitting of the training space, which would lead to a more discriminant representation for the cell image. The simple linear support vector machine is used for cell pattern identification because of its low computational cost, in particular for large-scale datasets. Experiments using the open HEp-2 cell dataset used in the ICIP2013 contest validate that the proposed strategy can achieve a much better performance than the widely used local binary pattern (LBP) histogram and its extensions, rotation invariant co-occurrence LBP, and pair wise rotation invariant co-occurrence LBP, and that the achieved recognition error rate is even very significantly below the observed intralaboratory variability.

      DOI: 10.1109/TBME.2014.2320294

      J-GLOBAL

      researchmap

    • Adaptive noise reduction and edge enhancement in medical images by using ICA 査読有り

      Xian-Hua Han, Yen-Wei Chen

      Computational Intelligence in Biomedical Imaging9781461472452   347 - 375   2014年7月1日

      詳細を見る

      記述言語:英語   掲載種別:論文集(書籍)内論文   出版者・発行元:Springer New York  

      This chapter focuses on the development of novel image enhancement and robust edge detection methods for practical medical image processing. It is known that the popular transformation-domain shrinkage approach for image enhancement applies a fixed mathematical basis to transform all images to be processed for noise or artifact reduction. However it is not adaptable to processed images, and then easily leads to blurring in the enhanced images. On the other hand, the techniques that are commonly used for edge detection are known as gradient and Laplacian operators (or mask), and smoothed gradient masks are typically used for edge detection in noisy images. However, these methods share a common major drawback wherein the associated masks are always fixed irrespective of the noise level in the images. In this study, we propose a novel learning-based method to adaptively deduce the transforming basis or masks from the processing data for medical image enhancement and robust edge detection. By using independent component analysis (ICA), the proposed learning-based method can extract suitable basis functions or masks for image transformation for processing data, which are adaptable to both the processed image and related noise in the image. The efficiency of the proposed learning-based method for medical image enhancement and edge detection is demonstrated experimentally using positron emission tomography (PET) and magnetic resonance imaging (MRI) medical images.

      DOI: 10.1007/978-1-4614-7245-2_13

      Scopus

      researchmap

    • Sparse representation for image super-resolution 査読有り

      Xian-Hua Han, Yen-Wei Chen

      Studies in Computational Intelligence552   123 - 150   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:Springer Verlag  

      This chapter concentrates the problem of recovery a high-resolution (HR) image from a single low-resolution input image. Recent research proposed to deal with the image super-resolution problem with sparse coding, which is based on the well reconstruction of any local image patch by a sparse linear combination of an appropriately chosen over-complete dictionary. Therein the chosen LR (Low-resolution) and HR (High-resolution) dictionaries have to be exactly corresponding for well reconstructing the local image patterns. However, the conventional sparse coding based image super-resolution usually achieves a global dictionary D=[D l
      D h ] by jointly training the concatenated LR and HR local image patches, and then reconstruct the LR and HR image as a linear combination of the separated D l and D h . This strategy only can achieve the global minimum reconstructing error of LR and HR local patches, and usually cannot obtain the exactly corresponding LR and HR dictionaries. In addition, the accurate coefficients for reconstructing the HR image patch using HR dictionary D h are also unable to be estimated using only the LR input and the LR dictionary D l . Therefore, this paper proposes to firstly learn the HR dictionary D h from the features of the training HR local patches, and then propagates the HR dictionary to the LR one, called as HR2LR dictionary propagation, by mathematical proving and statistical analysis. The effectiveness of the proposed HR2LR dictionary propagation in sparse coding for super-resolution is demonstrated by comparison with the conventional super-resolution approaches such as sparse coding and interpolation. © 2014 Springer-Verlag Berlin Heidelberg.

      DOI: 10.1007/978-3-642-54851-2_6

      Scopus

      researchmap

    • Robust isotropic super-resolution by maximizing a Laplace posterior for MRI volumes 査読有り

      Xian-Hua Han, Yutaro Iwamoto, Akihiko Shiino, Yen-Wei Chen

      Progress in Biomedical Optics and Imaging - Proceedings of SPIE9034   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE  

      Magnetic resonance imaging can only acquire volume data with finite resolution due to various factors. In particular, the resolution in one direction (such as the slice direction) is much lower than others (such as the in-plane direction), yielding un-realistic visualizations. This study explores to reconstruct MRI isotropic resolution volumes from three orthogonal scans. This proposed super- resolution reconstruction is formulated as a maximum a posterior (MAP) problem, which relies on the generation model of the acquired scans from the unknown high-resolution volumes. Generally, the deviation ensemble of the reconstructed high-resolution (HR) volume from the available LR ones in the MAP is represented as a Gaussian distribution, which usually results in some noise and artifacts in the reconstructed HR volume. Therefore, this paper investigates a robust super-resolution by formulating the deviation set as a Laplace distribution, which assumes sparsity in the deviation ensemble based on the possible insight of the appeared large values only around some unexpected regions. In addition, in order to achieve reliable HR MRI volume, we integrates the priors such as bilateral total variation (BTV) and non-local mean (NLM) into the proposed MAP framework for suppressing artifacts and enriching visual detail. We validate the proposed robust SR strategy using MRI mouse data with high-definition resolution in two direction and low-resolution in one direction, which are imaged in three orthogonal scans: axial, coronal and sagittal planes. Experiments verifies that the proposed strategy can achieve much better HR MRI volumes than the conventional MAP method even with very high-magnification factor: 10. © 2014 SPIE.

      DOI: 10.1117/12.2043361

      Scopus

      researchmap

    • Bayesian-based Saliency Model for Liver Tumor Enhancement 査読有り

      Xian-Hua Han, Yen-Wei Chen, Gang Xu

      SMART DIGITAL FUTURES 2014262   357 - +   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IOS PRESS  

      Automatic tumor enhancement and detection has an essential role for the computer-aided diagnosis of liver tumor in CT volume data. This paper proposes a novel tumor enhancement strategy by extracting a tumor saliency map, which represents the uncommon or tumor tissue compared to the liver and vessel ones in CT volumes. The saliency map can be constructed by exploring the existing probability of tumor in any voxel. However, the tumor prototypes in a test liver volume from a specific patient or common tumor prototypes are extremely difficult to achieve due to requirement of full-searching and large variation of tumor tissues in different liver volumes. Therefore, this paper investigates a tumor-training-data free strategy by only constructing the common healthy liver and vessel prototypes, which can be extracted from any slice of a liver volume, and then applies a non-parametric Bayesian framework for calculating the existing probability of liver or vessel. Finally, the existing probability of tumor can be deduced from that of liver or vessel. The advantages of our proposed strategy mainly include three aspects: (1) it only needs to construct the prototypes of common tissue such as liver or vessel region, which are easily obtained in any liver volume; (2) it proposes an adaptive non-parametric framework for tumor enhancement, which does not need to learn a common classification model for all liver volumes; (3) dispensable to remove the other different structure such as vessel in liver volume as a pre-processing step. Experiments validate that the proposed Bayesian-based saliency model for liver tumor enhancement can perform much better than the conventional approaches such as EM, EM/MPM tumor segmentation methods.

      DOI: 10.3233/978-1-61499-405-3-357

      researchmap

    • Fisher Vector of Micro-Texton for HEp-2 Staining Pattern Classification 査読有り

      Xian-Hua Han, Yen-Wei Chen

      IFAC PAPERSONLINE47 ( 3 ) 3575 - 3580   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:ELSEVIER SCIENCE BV  

      This study addresses the classification problem of HEp-2 cell using indirect immunofluorescent (IIF) image analysis, which can indicate the presence of autoimmune diseases by searching for antibodies in the patient serum. Generally, IIF analysis remains a subjective method, which depends too heavily on the experience and expertise of the physician. Recently, some studies show that it is possible to identify the cell patterns using IIF by image analysis and machine learning techniques. However, it still has large gap between automatic recognition and the physical experts' decision. This paper explores the discriminative feature extraction of HEp-2 cell images in IIF, and then identifies the patterns of HEp-2 cell using machine learning techniques. Motivated by the research progress on computer vision that small local pixel pattern distributions can be highly discriminative, the proposed strategy employs a parametric probability process to model the local image patches (Textons: micro structure in the cell image), and extract the higher-order statistics (also called Fisher-Vector) to the model parameters for the image description. The proposed strategy can adaptively characterize the micro-Texton space of HEp-2 cell images as the generative probability model, and learn the parameters for better fitting the training space, which would lead to more discriminant representation for the cell image. The simple linear support vector machine is combined for cell pattern identification due to its low computational cost especially for large-scale dataset. Experiments on the released HEp-2 cell dataset of ICIP2013 competition validate that the proposed strategy can achieve much better performance than the popular used local binary pattern (LBP) image descriptor, and the achieved recognition error rate is even greatly below the observed intra-laboratory variability.

      researchmap

    • Automatic Segmentation of Liver From CT Images Using Probabilistic Atlas and Template Matching 査読有り

      Yen-Wei Chen, Amir H. Foruzan, Chunhua Dong, Tomoko Tateyama, Xianhua Han

      SMART DIGITAL FUTURES 2014262   412 - 420   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IOS PRESS  

      A framework is proposed for automatic liver segmentation from CT volumes using probabilistic atlases and template matching techniques. Probabilistic atlases of human anatomy have been widely used for organ segmentation, which is used as a prior probability in a Bayes framework. The challenge is how to register the atlas to the patient volume. In this paper, we propose a template matching based technique for probabilistic atlas based organ segmentation. In our proposed method, we first find a Region of Interest (ROI) of the organ, which is based on human anatomic structure, and then the probabilistic atlas is used as a template to find the organ in the ROI by the use of template matching.

      DOI: 10.3233/978-1-61499-405-3-412

      researchmap

    • Alignment-Free and High-Frequency Compensation in Face Hallucination 査読有り

      Yen-Wei Chen, So Sasatani, Xian-Hua Han

      SCIENTIFIC WORLD JOURNAL   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:HINDAWI PUBLISHING CORPORATION  

      Face hallucination is one of learning-based super resolution techniques, which is focused on resolution enhancement of facial images. Though face hallucination is a powerful and useful technique, some detailed high-frequency components cannot be recovered. It also needs accurate alignment between training samples. In this paper, we propose a high-frequency compensation framework based on residual images for face hallucination method in order to improve the reconstruction performance. The basic idea of proposed framework is to reconstruct or estimate a residual image, which can be used to compensate the high-frequency components of the reconstructed high-resolution image. Three approaches based on our proposed framework are proposed. We also propose a patch-based alignment-free face hallucination. In the patch-based face hallucination, we first segment facial images into overlapping patches and construct training patch pairs. For an input low-resolution (LR) image, the overlapping patches are also used to obtain the corresponding high-resolution (HR) patches by face hallucination. The whole HR image can then be reconstructed by combining all of the HR patches. Experimental results show that the high-resolution images obtained using our proposed approaches can improve the quality of those obtained by conventional face hallucination method even if the training data set is unaligned.

      DOI: 10.1155/2014/903160

      researchmap

    • Robust Isotropic Super-Resolution by maximizing a Laplace posterior for MRI Volumes 査読有り

      Xian-Hua Han, Yutaro Iwamoto, Akihiko Shino, Yen-Wei Chen

      MEDICAL IMAGING 2014: IMAGE PROCESSING9034   90342I   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      Magnetic resonance imaging can only acquire volume data with finite resolution due to various factors. In particular, the resolution in one direction (such as the slice direction) is much lower than others (such as the in-plane direction), yielding un-realistic visualizations. This study explores to reconstruct MRI isotropic resolution volumes from three orthogonal scans. This proposed super-resolution reconstruction is formulated as a maximum a posterior (MAP) problem, which relies on the generation model of the acquired scans from the unknown high-resolution volumes. Generally, the deviation ensemble of the reconstructed high-resolution (HR) volume from the available LR ones in the MAP is represented as a Gaussian distribution, which usually results in some noise and artifacts in the reconstructed HR volume. Therefore, this paper investigates a robust super-resolution by formulating the deviation set as a Laplace distribution, which assumes sparsity in the deviation ensemble based on the possible insight of the appeared large values only around some unexpected regions. In addition, in order to achieve reliable HR MRI volume, we integrates the priors such as bilateral total variation (B TV) and non-local mean (NLM) into the proposed MAP framework for suppressing artifacts and enriching visual detail. We validate the proposed robust SR strategy using MRI mouse data with high-definition resolution in two direction and low-resolution in one direction, which are imaged in three orthogonal scans: axial, coronal and sagittal planes. Experiments verifies that the proposed strategy can achieve much better HR MRI volumes than the conventional MAP method even with very high-magnification factor: 10.

      DOI: 10.1117/12.2043361

      researchmap

    • Sparse and Low Rank Matrix Decomposition Based Local Morphological Analysis and Its Application to Diagnosis of Cirrhosis Livers 査読有り

      Junping Deng, Xianhua Han, Gang Xu, Yen-Wei Chen

      2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)   3363 - 3368   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      Cirrhosis liver is a terrible disease which is threatening our lives. Meanwhile, cirrhosis will cause significant hepatic morphological changes. While it is well known that the livers from different subjects have similar global shape structure which means liver shape ensemble should be low-rank. However the deformation which caused by cirrhosis can be considered as sparse compared with the whole liver. Therefore, in this study, we proposed to apply spare and low-rank matrix decomposition to partition the local deformation part (sparse error matrix E) from the global similar structure (low-rank matrix A) using the input liver shape D, which is the landmark coordinates of liver shapes and already have been aligned by the current rigid registration methods firstly. And then sparse matrix E is used for diagnosis. In common sense, the normal liver should have less local deformation than that of abnormal liver, which means that the norm of sparse matrix E for normal liver is smaller than the norm for abnormal one. Thus, we can simply use a threshold classify normal and abnormal livers using the norm of E for these two categories. The proposed method is evaluated by a liver database which includes 30 normal livers and 30 abnormal livers. The experimental results of proposed method is better than those of state of the art statistical shape model(SSM) based methods.

      DOI: 10.1109/ICPR.2014.579

      researchmap

    • Hybrid Aggregation of Sparse Coded Descriptors for Food Recognition 査読有り

      Riko Kusumoto, Xian-Hua Han, Yen-Wei Chen

      2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)   1490 - 1495   2014年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      Recent year, with the increasing of unhealthy diets which will threaten people's life due to the various resulted risks such as heart stroke, liver trouble and so on, the maintaining for healthy life has attracted much attention and then how to manage the dietary life is becoming more and more important. In this research, we aim to construct an auto-recognition system of food images and keep the daily food-log records which will contribute to manage dietary life. With the easily available food images taken by mobile phone, it prospects to give the insight about the daily dietary of users with our constructed food recognition system. In order to achieve the acceptable recognition performance of the food images, we propose to apply a sparse model for coding local descriptors extracted from the food images and various pooling methods for aggregating the xoded descriptors. Sparse coding: an extension of vector quantization for local descriptors, which is popularly used in Bag-of-Features (BoF) for image representation, can reconstruct the local descriptors more effective, and then obtain more discriminated feature for food image representation. However, in order to emphasize the strongest activated pattern, the widely applied aggregation strategy of the sparse coded vector is only to retain the maximum coefficient in all (named as Max-pooling), which would completely ignore the frequency: an important signature for identifying different types of images, of the activated patterns. Therefore, we explore a hybrid aggregation strategy named as top-ranked average pooling (TRAP), which integrates not only the maximum activated magnitude but also the stronger activated number for image representation. Experiments validate that the proposed hybrid aggregation strategy combined with sparse model can greatly improve the recognition rates compared with the conventional BOF model and the state-of-the-art methods on two databases: our constructed RFID and the public PFID.

      DOI: 10.1109/ICPR.2014.265

      researchmap

    • Generalized N-dimensional independent component analysis and its application to multiple feature selection and fusion for image classification

      Danni Ai, Guifang Duan, Xianhua Han, Yen-Wei Chen

      Neurocomputing103   186 - 197   2013年3月1日

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)  

      We propose a multilinear independent component analysis (ICA) framework called generalized N-dimensional ICA (GND-ICA) by extending the conventional linear ICA based on the multilinear algebra. Unlike the linear ICA that only treats one-dimensional data, the proposed GND-ICA treats N-dimensional data as a tensor without any preprocess of data vectorization. We furthermore introduce two types of GND-ICA solutions and analyze their efficiency and effectiveness. As an application, the GND-ICA can be used for multiple feature fusion and representation for color image classification. Many features extracted from a given image are constructed as a tensor. The feature tensor can be effective represented by GND-ICA. Compared with the conventional linear subspace learning methods, GND-ICA is capable of obtaining more distinctive representation for color image classification. © 2012 Elsevier B.V.

      DOI: 10.1016/j.neucom.2012.09.020

      Scopus

      researchmap

    • Global Shape Analysis of the Fibrotic Liver in CT Temporal Sequence

      Chen Yen-Wei, Luo Jie, Tateyama Tomoko, Han Xian-Hua, Furukawa Akira, Kanasaki Shuzo

      生体医工学51   M - 85-M-85   2013年

      詳細を見る

      記述言語:英語   出版者・発行元:Japanese Society for Medical and Biological Engineering  

      DOI: 10.11239/jsmbe.51.M-85

      CiNii Article

      researchmap

    • Automatic prediction of trait anxiety degree using recognition rates of facial emotions 査読有り

      Xinyin Huang, Dinye Chen, Yang Huang, Xianhua Han, Yen-Wei Chen

      2013 6th International Conference on Advanced Computational Intelligence, ICACI 2013 - Proceedings   272 - 275   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      The trait anxiety degree is a significant standard to measure the psychological status, and the measurement (score) of trait anxiety degree is generally obtained by a very complex text questionnaire, which usually takes large amount of time and is subjectively various according to environmental condition. On the other hand, the researches in psychological field have proven that personality recognition of different facial emotions is strongly related to the degree of trait anxiety. In this work, we propose to automatically predict the trait anxiety score using the recognition rates of different facial emotions. In order to select compact and discriminant features, we investigate a correlation-based feature selection strategy in both raw data and PCA transformed space. Experimental results show that our proposed strategy can achieve reasonable trait anxiety score, which, also can validates the reliable relation between recognition rates of facial emotion and trait anxiety score. © 2013 IEEE.

      DOI: 10.1109/ICACI.2013.6748515

      Scopus

      researchmap

    • Sparse dictionary representation and propagation for MRI volume super-resolution 査読有り

      Xian-Hua Han, Yen-Wei Chen

      Progress in Biomedical Optics and Imaging - Proceedings of SPIE8669   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      This study addresses the problem of generating a high-resolution (HR) MRI volume from a single low-resolution (LR) MRI input volume. Recent researches have proved that sparse coding can be successfully applied for single-frame super-resolution for natural images, which is based on good reconstruction of any local image patch with a sparse linear combination of atoms taken from an appropriate over-complete dictionary. This study adapts the basic idea of sparse code-based super-resolution (SCSR) for MRI volume data, and then improves the dictionary learning strategy in the conventional SCSR for achieving the precise sparse representation of HR volume patches. In the proposed MRI super-resolution strategy, we only learn the dictionary of the HR MRI volume patches with sparse coding algorithm, and then propagate the HR dictionary to the LR dictionary by mathematical analysis for calculating the sparse representation (coefficients) of any LR local input volume patch. The unknown corresponding HR volume patch can be reconstructed with the sparse coefficients from the LR volume patch and the corresponding HR dictionary. We validate that the proposed SCSR strategy through dictionary propagation can recover much clearer and more accurate HR MRI volume than the conventional interpolated methods. © 2013 SPIE.

      DOI: 10.1117/12.2008145

      Scopus

      researchmap

    • Computer-Aided Diagnosis and Quantification of Cirrhotic Livers Based on Morphological Analysis and Machine Learning 査読有り

      Yen-Wei Chen, Jie Luo, Chunhua Dong, Xianhua Han, Tomoko Tateyama, Akira Furukawa, Shuzo Kanasaki

      COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:HINDAWI LTD  

      It is widely known that morphological changes of the liver and the spleen occur during the clinical course of chronic liver diseases. In this paper, we proposed a morphological analysis method based on statistical shape models (SSMs) of the liver and spleen for computer-aided diagnosis and quantification of the chronic liver. We constructed not only the liver SSM but also the spleen SSM and a joint SSM of the liver and the spleen for a morphologic analysis of the cirrhotic liver in CT images. The effective modes are selected based on both its accumulation contribution rate and its correlation with doctor's opinions (stage labels). We then learn a mapping function between the selected mode and the stage of chronic liver. The mapping function was used for diagnosis and staging of chronic liver diseases.

      DOI: 10.1155/2013/264809

      researchmap

    • Quantifying Stage Progress of Cirrhotic Livers Based on Statistic Shape Models 査読有り

      Yen-Wei Chen, Chunhua Dong, Xian-Hua Han, Tomoko Tateyama, Shuzo Kanasaki, Akira Furukawa

      PROCEEDINGS OF THE 2013 6TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI 2013), VOLS 1 AND 2   822 - 825   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      It is widely known that morphological changes of the liver and the spleen occur during the clinical course of chronic liver diseases. In this paper, we show a preliminary study on quantifying stage progress of cirrhotic liver based on statistical shape models (SSMs). We constructed not only the liver SSM, but also the spleen SSM and a joint SSM of the liver and the spleen for a morphologic analysis of the cirrhotic liver in CT images. The effective modes are selected based on both its accumulation contribution rate and its correlation with doctor's opinions (stage labels). Both normal and abnormal livers including temporal sequence data are projected to the subspace with the selected modes. Most of the normal data are located in the center, while abnormal data are scattered around the normal data. If the disease progresses, data will move outside. The distance to the center can be used to quantify the stage progress of the cirrhotic liver.

      DOI: 10.1109/BMEI.2013.6747054

      researchmap

    • 肝臓の統計形状モデル構築と肝硬変症支援診断システムへの応用

      上谷 芽衣, 健山 智子, 小原 伸哉, 田中 英俊, 韓 先花, 金﨑 周造, 古川 顕, 陳 延偉

      電気学会論文誌. C133 ( 11 ) 2037 - 2043   2013年

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人 電気学会  

      In recent years, there are increasing interests in statistical shape modeling of human anatomy. The statistical shape model can capture the morphological variations of human anatomy. Since the liver cirrhosis will cause significant morphological changes, the authors propose a computer-aided diagnosis method for liver cirrhosis based on statistical shape models. In the proposed method, the authors first construct a statistical shape model of the liver using 50 clinical CT datasets (25 sets of normal data and 25 sets of abnormal data). The authors apply marching cubes algorithm to convert the segmented liver volume to a triangulated mesh surface which containing 1000 vertex points. The coordinates of these vertex points are used to represent 3D liver shape as a shape vector. After normalization and correspondence finding between all datasets, Principal Component Analysis (PCA) is employed to find the principal variation modes of shape vectors. Then the authors propose a mode selection method based on class variations between the normal class and abnormal class. The authors found the top two modes of class variations are most effective for classification of normal and abnormal livers. The classification rate of abnormal livers and normal liver are 84% and 80%, respectively, by the use of a simple linear discriminant function.

      DOI: 10.1541/ieejeiss.133.2037

      CiNii Article

      researchmap

    • Application of Statistical Shape Model of the Liver in Classification of Cirrhosis

      Yen-Wei Chen, Mei Uetani, Shinya Kohara, Tomoko Tateyama, Xian-Hua Han, Akira Furukawa, Shuzo Kanasaki

      International Journal of Digital Content Technology and its Applications7 ( 9 ) 477-484   2013年

      詳細を見る

    • 患者に特化した肝臓情報とその脈管分布可視化および対話かつ直感的な手術支援システムの構築

      健山 智子, 韓 先花, 姜 彗研, 陳 延偉, 海堀 昌樹, 進藤 典, Amir Hossein FORUZAN, 林 正倫, 宮脇 康介, 津田 匠, 松井 康輔, 權 雅憲

      Medical Imaging Technology31 ( 3 ) 176 - 188   2013年

      詳細を見る

      記述言語:日本語   出版者・発行元:日本医用画像工学会  

      肝臓は個人においてもその大きさ,形状および脈管分布においてバリエーションが広く,手術対象患者の肝臓に対応した手術支援は重要な課題である.本論文では手術対象となる特定患者の腹部 CT画像を用いてその患者に特化した肝臓情報を用いて擬似手術環境を再現することができる手術支援システムを構築する.提案するシステムは大きく 3つのプロセス,(1)画像処理技術による肝臓領域分割とその内部脈管分布の半自動抽出,(2)コンピュータビジョン技術による対象患者の肝臓三次元可視化による擬似手術環境構築,(3)構築された擬似手術環境上で仮想メス操作による対話的かつ直感的な切離ラインの再現によって開発する.本研究で構築したシステムの有効性を確認するため,臨床外科医が実際の症例にもとづいて本システム上で対話的に仮想肝臓内を侵襲しながら切離ラインを設定するシミュレーションを行った.本システム上で操作を行った臨床医に対しシミュレーション評価を客観評価および主観評価を行ったところ,事前の手術支援として有効なシステムであると評価を得た.

      DOI: 10.11409/mit.31.176

      CiNii Article

      researchmap

    • Gradient-based edge preserving interpolation and its application to super-resolution 査読有り

      Yutaro Iwamoto, Xian-Hua Han, Tomoko Tateyama, Motonori Ohashi, So Sasatani, Yen-Wei Chen

      ELECTRONICS AND COMMUNICATIONS IN JAPAN96 ( 1 ) 43 - 50   2013年1月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:WILEY-BLACKWELL  

      Super-resolution is a process for obtaining high-quality, high-resolution images from one or a set of low-resolution images. The most practical methods for image super-resolution are reconstruction-based methods, which minimize the difference between observed low-resolution images and the estimate for high-resolution images. Therein, the interpolation step plays a key role in the estimated high-resolution image quality. Usually, the conventional bilinear or bicubic methods are used in reconstruction-based super-resolution. However, these conventional interpolations generally lead to blurring in edge regions and need more time for convergence in the reconstruction-based super-resolution method. Therefore, this paper proposes a gradient-based edge-preserving interpolation method, which can reduce not only artifact noise but also blurring near the edge regions in the estimated high-resolution image. Furthermore, our proposed interpolation method can also solve high-complexity, time-consuming problems in the recently developed new edge-directed interpolation, which usually can achieve sharp edges in the high-resolution reconstructed image. Experiments confirm that our proposed interpolation method for image super-resolution is more effective than the conventional interpolation methods. (c) 2012 Wiley Periodicals, Inc. Electron Comm Jpn, 96(1): 4350, 2013; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ecj.11413

      DOI: 10.1002/ecj.11413

      researchmap

    • Global and local features for accurate impression estimation of cloth fabric images. 査読有り

      Xinyin Huang, Dingye Chen, Xian-Hua Han, Yen-Wei Chen

      Proceedings of the 2013 IEEE/SICE International Symposium on System Integration, SII 2013, Kobe, Japan, December 15-17, 2013   486 - 489   2013年

      詳細を見る

      出版者・発行元:IEEE  

      DOI: 10.1109/SII.2013.6776758

      researchmap

    • Food Recognition Using Codebook-based Model with Sparse-Coding 査読有り

      Minami Wazumi, Xian-Hua Han, Yen-Wei Chen

      2013 IEEE/SICE INTERNATIONAL SYMPOSIUM ON SYSTEM INTEGRATION (SII)   482 - 485   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Recently, with the increasing of unhealthy diets and the attracted attention for healthy life, how to manage the dietary life is becoming more and more important. In this paper, we aim to construct a food-log system, which can auto-recognize the menu contents from food image taken by mobile phone. In order to increase recognition rate of food image, this research explores the typical codebook-based mode-Bag-of Feature (BOF), and the improved sparse-coding (Sc) based one for image representation, which is one of the main factors for affecting on food-log system. Furthermore, instead of pooling all the coded local features globally, the Spatial Pyramid Matching (SPM) strategy is adopted to integrated them from various spatial scale sub-regions. The experimental results validate that the Sc-based codebook model can achieve much better performance than the typically Bag-of-Feature on dish recognition.

      DOI: 10.1109/SII.2013.6776730

      researchmap

    • Generalized n-dimensional PCA for compact representation of medical volumes and effective mode selection by adaboost 査読有り

      Junping Deng, Rui Xu, Xianhua Han, Yen-Wei Chen

      Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)8294   629 - 636   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer Verlag  

      The paper addresses the issues of compact representation of medical volumes and effective mode selection. Medical volumes are unlike traditional two dimensional images in pattern recognition which in spatial domain has three dimensions. Meanwhile it often meets small sample problem. Thus common compact representation methods, such as PCA, do not fit medical volumes. Because the number of eigenvector is so little that it will lose much useful information. In previous work, we proposed a Generalized N-dimensional principal component analysis (GND-PCA) for reconstruction of medical volumes with only few samples. The core tensor of GND-PCA can keep most useful information. However, making diagnosis using the core tensor is difficult due to most modes very general for all samples. It will affect finally diagnosis. To resolve this problem, Adaboost is used as classifier in diagnosis because it can choose distinctive mode according to its definition. The proposed method was evaluated using a medical volume database. In our experiment, we compare Adaboost with SVM and KNN. The classification accuracy of Adaboost is slightly better than that of SVM and KNN,meanwhile time consuming of classification of Adaboost is greatly less than those of other two classification methods. © Springer International Publishing Switzerland 2013.

      DOI: 10.1007/978-3-319-03731-8_58

      Scopus

      researchmap

    • Sparse Dictionary Representation and Propagation for MRI Volume Super-Resolution 査読有り

      Xian-Hua Han, Yen-Wei Chen

      MEDICAL IMAGING 2013: IMAGE PROCESSING8669   86692N   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      This study addresses the problem of generating a high-resolution (HR) MRI volume from a single low-resolution (LR) MRI input volume. Recent researches have proved that sparse coding can be successfully applied for single-frame super-resolution for natural images, which is based on good reconstruction of any local image patch with a sparse linear combination of atoms taken from an appropriate over-complete dictionary. This study adapts the basic idea of sparse code-based super-resolution (SCSR) for MRI volume data, and then improves the dictionary learning strategy in the conventional SCSR for achieving the precise sparse representation of HR volume patches. In the proposed MRI super-resolution strategy, we only learn the dictionary of the HR MRI volume patches with sparse coding algorithm, and then propagate the HR dictionary to the LR dictionary by mathematical analysis for calculating the sparse representation (coefficients) of any LR local input volume patch. The unknown corresponding HR volume patch can be reconstructed with the sparse coefficients from the LR volume patch and the corresponding HR dictionary. We validate that the proposed SCSR strategy through dictionary propagation can recover much clearer and more accurate HR MRI volume than the conventional interpolated methods.

      DOI: 10.1117/12.2008145

      researchmap

    • Residual image compensations for enhancement of high-frequency components in face hallucination 査読有り

      Yen-Wei Chen, So Sasatani, Xianhua Han

      Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)7951 ( 1 ) 627 - 634   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:Springer  

      Recently a growing interest has been seen in single-frame super-resolution techniques, which are known as example-based or learning based super-resolution techniques. Face Hallucination is one of such techniques, which is focused on resolution enhancement of facial images. Though face hallucination is a powerful and useful technique, some detailed high-frequency components cannot be recovered. In this paper, we propose a high-frequency compensation framework based on residual images for face hallucination method in order to improve the reconstruction performance. The basic idea of proposed framework is to reconstruct or estimate a residual image, which can be used to compensate the high-frequency components of the reconstructed high-resolution image. Three approaches based on our proposed framework are proposed. Experimental results show that the high-resolution images obtained using our proposed approaches can improve the quality of those obtained by conventional face hallucination method. © 2013 Springer-Verlag Berlin Heidelberg.

      DOI: 10.1007/978-3-642-39065-4-75

      Scopus

      researchmap

    • PILOT STUDY OF APPLYING SHAPE ANALYSIS TO LIVER CIRRHOSIS DIAGNOSIS 査読有り

      Jie Luo, Yen-Wei Chen, Xian-Hua Han, Tomoko Tateyama, Akira Furukawa, Shuzo Kanasaki

      2013 20TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP 2013)   3537 - 3541   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper explores the potential of applying shape analysis to classify normal/cirrhotic liver and in addition estimate the severity of abnormal cases. Conventional Computer-Aided Diagnosis (CAD) systems are developed for automatically providing a binary output as a second opinion to assist radiologists to draw conclusions about the condition of the pathology (normal or abnormal). After the disease is diagnosed, grasping the proceeding stage of the abnormal degree is essential for adopting the appropriate strength of treatment. However, none of existing CAD system is well established for such a challenging task. Liver cirrhosis has an important feature: morphological changes of the liver and the spleen occur during the clinical course of liver cirrhosis. In this study we constructed liver, spleen and their joint Statistical Shape Models (SSMs) to quantitatively assess the global shape variation and selected several modes from the SSMs. Then we learnt a mapping function between coefficients of selected modes and the ground truth staging label by Support Vector Regression (SVR). Using this mapping function, the proceeding stage of new input data can be estimated. Experimental results have validated the potential of our method on assisting the cirrhosis diagnosis.

      DOI: 10.1109/ICIP.2013.6738730

      researchmap

    • A Morphologic Analysis of Cirrhotic Liver in CT Images 査読有り

      Yen-Wei Chen, Jie Luo, Xianhua Han, Tomoko Tateyama, Akira Furukawa, Shuzo Kanasaki

      IMAGE ANALYSIS AND RECOGNITION7950   494 - 501   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      Cirrhosis will cause significant morphological changes on both liver and spleen. In this paper, we constructed not only the liver statistical shape models (SSM), but also the spleen SSM and a joint SSM of the liver and the spleen for a morphologic analysis of the cirrhotic liver in CT images. We also proposed a mode selection method based on both its accumulation contribution rate and its correlation with doctor's opinions (labels). The classification performance for normal and abnormal livers is significantly improved by our proposed method. The classification accuracies for normal and cirrhotic livers are 88% and 90%, respectively.

      DOI: 10.1007/978-3-642-39094-4_56

      researchmap

    • Computer-aided liver surgical planning system using CT volumes. 査読有り

      Yen-Wei Chen, Masaki Kaibori, Tsukasa Shindo, Kousuke Miyawaki, Amir Hossein Foruzan, Tomoko Tateyama, Xian-Hua Han, Kosuke Matsui, Takumi Tsuda, A.-Hon Kwon

      35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2013, Osaka, Japan, July 3-7, 2013   2360 - 2363   2013年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/EMBC.2013.6610012

      researchmap

    • Sparse Model in Hierarchic Spatial Structure for Food Image Recognition 査読有り

      Riko Kusumoto, Xian-Hua Han, Yen-Wei Chen

      PROCEEDINGS OF THE 2013 6TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI 2013), VOLS 1 AND 2   851 - 855   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Recent year, with the increasing of unhealthy diets which will threaten people's life due to the various resulted risks such as heart stroke, liver trouble and so on, the remain for healthy life has attracted much attention and then how to manage the dietary life is becoming more and more important. In this research, we aim to construct a auto-recognition system of food images and keep the daily food-log records which will contribute to manage dietary life. With the easily available food images taken by mobile phone, it prospects to give the insight about the daily dietary of users with our constructed food recognition system. In order to achieve the acceptable recognition performance of the food images, we propose to apply a sparse model for coding a local descriptor extracted from the food images. Sparse coding: an extension of vector quantization for local descriptors, which is popularly used in Bag-of-Features (BoF) for image representation in generic object recognition, can represent the local descriptors more efficient, and then abtain more discriminant feature for food image representation. Moreover, in order to introduce spatial information, a hierarchic spatial structure is explored to extract the feature based sparse model. Experiments validate that the proposed strategy can greatly improve the recognition rates compared with the conventional BOF model on two databases: our constructed RFID and the public PFID.

      DOI: 10.1109/BMEI.2013.6747060

      researchmap

    • Robust Local Ternary Patterns for Texture Categorization 査読有り

      Xian-Hua Han, Gang Xu, Yen-Wei Chen

      PROCEEDINGS OF THE 2013 6TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI 2013), VOLS 1 AND 2   846 - 850   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper proposes a new image representation for texture categorization, which is based on extension of local binary patterns (LBP). As we know LBP can achieve effective description ability with appearance invariance and adaptability of patch matching based methods. However, LBP only thresholds the differential values between neighborhood pixels and the focused one to 0 or 1, which is very sensitive to noise existing in the processed image. This study extends LBP to local ternary patterns (LTP), which considers the differential values between neighborhood pixels and the focused one as negative or positive stimulus if the absolute differential value is large; otherwise no stimulus (set as 0). With the ternary values of all neighbored pixels, we can achieve a pattern index for each local patch, and then extract the pattern histogram for image representation. Experiments on two texture datasets: Brodats32 and KTH TIPS2-a validate that the robust LTP can achieve much better performances than the conventional LBP and the state-of-the-art methods.

      DOI: 10.1109/BMEI.2013.6747059

      researchmap

    • High Frequency Compensated Face Hallucination with Total Variation Constraint 査読有り

      Yusuke Nojima, Xian-Hua Han, Kazuki Taniguchi, Yen-Wei Chen

      PROCEEDINGS OF THE 2013 6TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI 2013), VOLS 1 AND 2   831 - 835   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Face Hallucination, one of the most famous learning-based image super-resolution techniques for facial images, can reconstruct a high-resolution image using only one low-resolution image. However, some detailed high-frequency components of the reconstructed image cannot be recovered using this method. In addition, the available LR images are sometimes blurred because of object movement or hardware problems. In this study, we proposed a high-frequency compensated face hallucination method with total variation constraint for HR image recovery. The proposed method is divided into two main processes: 1) Deblurring the LR input with total variation constraint; 2) Super-resolution with the proposed high-frequency compensated face hallucination. Experimental results show that the highresolution images obtained by our proposed approach are much better than those obtained by conventional methods.

      DOI: 10.1109/BMEI.2013.6747056

      researchmap

    • ADAPTIVE COLOR DISCRIMINATION FOR IMAGE CLASSIFICATION 査読有り

      Motoki Nakajima, Yen-Wei Chen, Xian-Hua Han

      PROCEEDINGS OF THE 2013 6TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI 2013), VOLS 1 AND 2   826 - 830   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Semantic understanding of images remains an important research challenge in machine intelligence and statistical learning. It mainly includes two steps: feature extraction and classification. This study mainly focuses scene image recognition, where color information plays an important role. The conventional color representation of images mainly includes color distribution (histogram) and its statistical information based on uniformed quantization color bin. However, for a specific recognition application such several scene types, some quantized colors maybe never appear in any scene image, and at the same time the detail variation in other quantized colors include much discriminative features. Therefore, this study proposes to characterize the color information of scene images using a leaning strategy for producing adaptive color level, and extract the histogram of the learned color levels for image representation. With the proposed strategy, the compact (learned) color levels can represent the image in our application more faithful than the uniform quantized conventional RGB color. Experimental results show that the recognition rate with our proposed methods can be greatly improved compared to the conventional color histogram.

      DOI: 10.1109/BMEI.2013.6747055

      researchmap

    • Nonrigid Registration for Evaluating Locoregional Therapy of Hepatocellular Carcinoma 査読有り

      Chunhua Dong, Toshihito Seki, Ryosuke Inoguchi, Chen-Lun Lin, Xianhua Han, Yen-Wei Chen

      PROCEEDINGS OF THE 2013 6TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI 2013), VOLS 1 AND 2   811 - 816   2013年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      The assessment of the treated margin with locoregional therapy (LT), for hepatocellular carcinoma (HCC), is the common method for predicting HCC recurrence in most hospital. However, tumors sometimes cannot be removed clearly with LT in limited conditions. The therapeutic efficiency of HCC is often evaluated by comparing 2D fusion images of computed tomography (CT) or magnetic resonance imaging (MRI) between the preoperation and the postoperation. However, judgment about whether the tumors exist in the treated margin after LT by using 2D slices sometimes is difficult. It is desirable to develop a suitable image registration algorithm to automatically align the two volumes in order to transform the treated margin of the postoperative volume to the tumor of the preoperative volume to assess the therapeutic efficiency after treatment of HCC. With taking these into consideration, this paper proposed an automatic 3D fusion imaging approach for medical image by using the nonrigid registration method that aligning an ablative marginthat is the treated margin after LT, onto the locations of HCC. In our registration algorithm, a rigid global transformation combined with localized B-spline is used to estimate the significant nonrigid motions of the liver between before and after LT. Our proposed approach can ensure the feasibility, the accuracy and the efficacy to assess the treated margin for HCC. Furthermore, this method can be adapted to register multimodality medical images. We demonstrate the effectiveness of our proposed method by comparing the difference criterions of fusion evaluation on medical images. The results clearly indicate that our method extremely useful in the evaluation of the treated margin, in addition, it remain the motion and local deformation of the volume.

      DOI: 10.1109/BMEI.2013.6747052

      researchmap

    • A preliminary study on multi-touch based medical image analysis and visualization system. 査読有り

      Jian Wang 0004, Hua-Wei Tu, Xian-Hua Han, Tomoko Tateyama, Yen-Wei Chen

      6th International Conference on Biomedical Engineering and Informatics, BMEI 2013, Hangzhou, China, December 16-18, 2013   797 - 801   2013年

      詳細を見る

      掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      DOI: 10.1109/BMEI.2013.6747049

      researchmap

    • Computer-Aided Diagnosis and Quantification of Cirrhotic Livers Based on Morphological Analysis and Machine Learning.

      Yen-Wei Chen, Jie Luo, Chunhua Dong, Xian-Hua Han, Tomoko Tateyama, Akira Furukawa, Shuzo Kanasaki

      Comput. Math. Methods Medicine2013 ( ** ) 264809 - 8   2013年

      詳細を見る

      掲載種別:研究論文(学術雑誌)  

      DOI: 10.1155/2013/264809

      researchmap

    • 多様体学習法と事例ベース法を融合した超解像技術とその高速化

      谷口 和輝, 韓 先花, 大橋 基範, 岩本 祐太郎, 笹谷 聡, 陳 延偉

      電気学会論文誌. C, 電子・情報・システム部門誌 = The transactions of the Institute of Electrical Engineers of Japan. C, A publication of Electronics, Information and Systems Society132 ( 11 ) 1768 - 1773   2012年11月1日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人 電気学会  

      This paper presents a new method for single-frame Super-Resolution (SR), by combining Example-based SR and neighbor embedding based SR (NE-based SR). Example-based SR attempts to generate High-Resolution (HR) image through estimating the High-Frequency (HF) components that are lost in the input Low-Resolution (LR) image. This method usually can achieve acceptable HR images if enough amounts of similar training samples are prepared. However, the HF component is approximated by only one training sample, which easily produces noise and artifacts. On the other hand, NE-based SR recovers HR image using manifold learning - Locally Linear Embedding, which represents any LR input and its corresponding HR one by a weighted linear combination of several training patches. The NE-based SR need to prepare large-scale training database with both intensity and structure variation, which will lead to high computation. This study combines these two methods to only estimate the HF components using several training samples. Moreover, we extend the proposed method to a fast version by processing only the patches with large variance. Experimental results show that the reconstructed HR images by our proposed approach are much better than those by conventional methods and interpolation techniques, and at the same time the computation is much faster.

      DOI: 10.1541/ieejeiss.132.1768

      Scopus

      CiNii Article

      researchmap

    • Multiple Watermarks for Management in Medical Image based on DWT 査読有り

      Chunhua Dong, Jingbing Li, Xianhua Han, Yen-Wei Chen

      International Journal of Digital Content Technology and its Applications6 ( 10 ) 239 - 247   2012年6月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)  

      Medical images must be stored and translated in a secure way. This paper proposes a multiple watermarks technique based on medical image features to serve these purposes. A part of sign sequence of DWT and DFT coefficients are used as feature vector for enhancing the robustness against rotation, scaling, translation and cropping attacks. In contrast to conventional approaches, our method can strengthen the fidelity of images through considering local image characteristics, without sacrificing the quality of medical images. We describe how to exact the feature vector of medical image, and then, embed and extract the multiple watermarks. The simulation results demonstrate that the watermarking scheme has strong robustness, and can embed much more data compared with the existing watermarking techniques.

      DOI: 10.4156/jdcta.vol6.issue10.28

      Scopus

      researchmap

    • View-Based Object Recognition Using ND Tensor Supervised Neighborhood Embedding 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE95D ( 3 ) 835 - 843   2012年3月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      In this paper, we propose N-Dimensional (ND) Tensor Supervised Neighborhood Embedding (ND TSNE) for discriminant feature representation, which is used for view-based object recognition. ND TSNE uses a general Nth order tensor discriminant and neighborhood-embedding analysis approach for object representation. The benefits of ND TSNE include: (1) a natural way of representing data without losing 'structure information, i.e., the information about the relative positions of pixels or regions; (2) a reduction in the small sample size problem, which occurs in conventional supervised learning because the number of training samples is much less than the dimensionality of the feature space; (3) preserving a neighborhood structure in tensor feature space for object recognition and a good convergence properly in training procedure. With Tensor-subspace features, the random forests is used as a multi-way classifier for object recognition, which is much easier for training and testing compared with multi-way SVM. We demonstrate the performance advantages of our proposed approach over existing techniques using experiments on the COIL-100 and the ETH-80 datasets.

      DOI: 10.1587/transinf.E95.D.835

      J-GLOBAL

      researchmap

    • Multilinear Supervised Neighborhood Embedding of a Local Descriptor Tensor for Scene/Object Recognition 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      IEEE TRANSACTIONS ON IMAGE PROCESSING21 ( 3 ) 1314 - 1326   2012年3月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC  

      In this paper, we propose to represent an image as a local descriptor tensor and use a multilinear supervised neighborhood embedding (MSNE) for discriminant feature extraction, which is able to be used for subject or scene recognition. The contributions of this paper include: 1) a novel feature extraction approach denoted as the histogram of orientation weighted with a normalized gradient (NHOG) for local region representation, which is robust to large illumination variation in an image; 2) an image representation framework denoted as the local descriptor tensor, which can effectively combine a moderate amount of local features together for image representation and be more efficient than the popular existing bag-of-feature model; and 3) an MSNE analysis algorithm, which can directly deal with the local descriptor tensor for extracting discriminant and compact features and, at the same time, preserve neighborhood structure in tensor-feature space for subject/scene recognition. We demonstrate the performance advantages of our proposed approach over existing techniques on different types of benchmark database such as a scene data set (i.e., OT8), face data sets (i.e., YALE and PIE), and view-based object data sets (COIL-100 and ETH-80).

      DOI: 10.1109/TIP.2011.2168417

      J-GLOBAL

      researchmap

    • Example-Based Super-Resolution using Locally Linear Embedding 査読有り

      Kazuki Taniguchi, Motonori Ohashi, Xian-Hua Han, Yutaro Iwamoto, So Sasatani, Yen-Wei Chen

      2011 6TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND CONVERGENCE INFORMATION TECHNOLOGY (ICCIT)   861 - 865   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Example-Based Super-Resolution is a learning-based technique that attempts to recover high-resolution (HR) image according to the corresponding relation in a set of training low-resolution (LR) and high-resolution image pairs prepared in advance. The conventional learning-based method for image super-resolution usually cannot achieve the high-frequency components accurately, which are lost in the input LR image, for recovering the HR image, since it only estimates the lost information using one most similar training LR patch to the input patch, and its corresponding HR pair. Therefore, we propose to use a manifold learning method- Locally Linear Embedding (LLE) for reconstructing the input LR patch with a linear weight summation of its several most similar training LR patches, and then can recover HR patch using the same linear summation of the corresponding training HR patches. Furthermore, in order to solve the expensive computational problem in the conventional exampled-based learning method, only the patches with larger variance, which means with high-frequency components, are selected for super-resolution procedures. Finally, Experimental results show that the recovered high-resolution images by our proposed approach are much better than those by conventional method and interpolation techniques.

      researchmap

    • Auto-Recognition of Food Images Using SPIN Feature for Food-Log System 査読有り

      Minami Wazumi, Xian-Hua Han, Danni Ai, Yen-Wei Chen

      2011 6TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND CONVERGENCE INFORMATION TECHNOLOGY (ICCIT)   874 - 877   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Recently, with the increasing of unhealthy diets and the attracted attention for healthy life, how to manage the dietary life is becoming more and more important. In this paper, we aim to construct a system, which can auto-recognize the menu contents from food image taken by mobile phone. As we know that the viewpoints can be varied in any direction when taking food images, and then, rotation-robust features for image representation are very important. Therefore, in this paper, we propose to extract rotation invariant features using circle-segmentation called SPIN for food recognition, and construct a Food-Log system, which records the contents of food menu, calories and nutritional value for management of the dietary life.

      researchmap

    • DFT Based Multiple Watermarks for Medical Image Robust to Common and Geometrical Attacks 査読有り

      Jingbing Li, Chunhua Dong, Xianhua Han, Yen-wei Chen

      2012 6TH INTERNATIONAL CONFERENCE ON NEW TRENDS IN INFORMATION SCIENCE, SERVICE SCIENCE AND DATA MINING (ISSDM2012)   472 - 477   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper presents a method for robust watermarking of medical image using DFT. A feature vector of the medical image is utilized to enhance the robustness against rotation, scaling, and translation. The proposed algorithm utilizes the medical image's feature, Hash function, the third party authentication. We describe how to obtain the feature vector of medical image and embed and extract the watermarking. Simulation results demonstrate the proposed algorithm's robustness against common and geometrical attacks.

      researchmap

    • Robust Multiple Watermarks for Medical Image Based on DWT and DFT 査読有り

      Jingbing Li, Chunhua Dong, Xianhua Han, Yen-wei Chen

      2011 6TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND CONVERGENCE INFORMATION TECHNOLOGY (ICCIT)   895 - 899   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Medical images must be stored in a secure way to preserve unauthorized disclosure of patient data. This paper proposes an algorithm of robust multiple watermarks to serve these purposes using DWT and DFT. A part of sign sequence of DWT and DFT coefficients are used as feature vector for enhancing the robustness against rotation attacks, scaling attacks, translation attacks and cropping attacks. Moreover, the content of medical image remains unchanged with our proposed algorithm which is one kind of the zero-watermarking technology. We describe how to extract the feature vector of medical image, and then, embed and extract the multiple watermarks. The results of experiment indicate that the watermark scheme has strong robustness, and can embed much more data compared with the existing watermarking techniques.

      researchmap

    • A New Linear Coding Algorithm for Efficient Multi-dimensional Data Representation without Data Expansion 査読有り

      Xu Qiao, Xuantao Su, Xianhua Han, Yen-Wei Chen

      2012 6TH INTERNATIONAL CONFERENCE ON NEW TRENDS IN INFORMATION SCIENCE, SERVICE SCIENCE AND DATA MINING (ISSDM2012)   478 - 482   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Linear coding is used for finding succinct representations of data sets. It also discover basis functions that capture higher-level features in the data. However, finding linear codes for multi-dimensional data remains a very difficult computational problem. Motivated by the work of linear image coding and multilinear algebra, we propose a linear tensor coding algorithm (LTC), which is applied to represent multi-dimensional data succinctly by a linear combination of tensor-formed bases without data expansion. Each basis captures some specific variability. The coefficients of data, which are associated with the bases, can be applied for representation, compression and classification. When we applied LTC algorithm on the phantom data, experimental results illustrate that our algorithm not only produces localized bases but also preserve the information of the input data.

      researchmap

    • Super-Resolution of Medical Volumes Based on Principal Component Regression 査読有り

      Yutaro Iwamoto, Xian-Hua Han, So Sasatani, Kazuki Taniguchi, Yen-Wei Chen

      2011 6TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND CONVERGENCE INFORMATION TECHNOLOGY (ICCIT)   945 - 948   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In medical imaging, the data resolution is usually insufficient for accurate diagnosis in clinical medicine. Especially in most case, the resolution in the slice direction (Z direction) is much lower than that of the in-plane resolution (XY direction). Therefore it is difficult to construct isotropic voxels, which is very important in 3-D visualization systems, such as surgical system. In this paper, we propose a method for improving resolution in the slice direction for medical volume images based on Principal Component Regression (PCR), which can be considered as one of the learning based super-resolution techniques. The experimental results verify the effectiveness of the proposed method by comparison with the conventional interpolation methods.

      researchmap

    • Statistical Shape Model of the Liver and Effective Mode Selection for Classification of Liver Cirrhosis 査読有り

      Yen-Wei Chen, Jie Luo, Tomoko Tateyama, Xian-Hua Han, Akira Furukawa, Shuzo Kanasaki, Huiyan Jiang

      2012 6TH INTERNATIONAL CONFERENCE ON NEW TRENDS IN INFORMATION SCIENCE, SERVICE SCIENCE AND DATA MINING (ISSDM2012)   449 - 452   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In computational anatomy, statistical shape model is used for quantitative evaluation of the variations of an organ shape. Since liver cirrhosis will cause significant hepatic morphological changes, we applied statistical shape model of the liver to capture the morphological changes and recognize whether a liver is normal or abnormal. In this paper, we propose an effective mode selection method to improve the classification accuracy. In addition to the conventional Accumulated Variance Contribution Rate (AVCR) based mode selection, we newly propose a Pearson correlation based mode selection method and combine them to select the effective modes. The coefficients of the selected modes (components) are used as features to recognize whether liver is normal or abnormal. The effectiveness of the proposed method is evaluated by the classification accuracy of normal and abnormal. Experimental results show that our proposed method is superior than conventional methods.

      researchmap

    • Diagnosis of liver cirrhosis with the use of multi-detector row computed tomography (MDCT): morphological approach and quantitative approach using statistical geometric hepatic model 査読有り

      Shuzo Kanasaki, Akira Furukawa, Makoto Wakamiya, Kiyoshi Murata, Shinya Kohara, Tomoko Tateyama, Xian-hua Han, Yen-Wei Chen

      2011 6TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND CONVERGENCE INFORMATION TECHNOLOGY (ICCIT)   959 - 962   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In clinical course of the chronic liver disease, the liver finally develops liver cirrhosis from the state of the chronic hepatitis. In the process, the patients are complicated with esophagogastric varices due to the portal hypertension and a hepatocellular carcinoma. It is very useful to be diagnosed using MDCT morphologically. A quantitative assessment has been tried from the CT images of the liver. A morphologic diagnosis of the chronic liver disease using the CT and a quantitative assessment are described in this article.

      researchmap

    • Shape Representation of Human Anatomy using Spherical Harmonic Basis Function 査読有り

      Tomoko Tateyama, Megumi Okegawa, Shinya Kohara, Xianhua Han, Shuzo Kanasaki, Akira Furukawa, Huiyan Jiang, Kiyoshi Murata, Yen-Wei Chen

      2011 6TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND CONVERGENCE INFORMATION TECHNOLOGY (ICCIT)   866 - 869   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In medical imaging research, the three-dimensional (3-D) shape representation and analysis of anatomic structures using only a few parameters is an important issues, and can be applied to computer assisted diagnosis, surgical simulations, visualization, and many other medical applications. This paper proposes the representation of the shape surfaces of a simply connected 3-D object using spherical harmonic (spharm) functions, which can provide an approximate global description of the shape using only a few spharm parameters. Spharms are obtained from several partial differential equations used in physics, such as the Laplace, Helmholtz, and Schrodinger equations, in which spherical coordinates are used. In this study, we aim to develop a 3D representation of human anatomy, such as the liver and spleen, with fewer parameters by using spharm functions. Furthermore, the experimental results show that the shape of the liver and spleen can be accurately represented with only a few spharm parameters.

      researchmap

    • Face Recognition Using Multilinear Manifold Analysis of Local Descriptors 査読有り

      Xian-Hua Han, Yen-Wei Chen

      STRUCTURAL, SYNTACTIC, AND STATISTICAL PATTERN RECOGNITION7626   734 - 742   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      In this paper, we propose to represent a face image as a local descriptor tensor and use a Multilinear Manifold Analysis (MMA) method for discriminant feature extraction, which is used for face recognition. The local descriptor tensor, which is a combination of the descriptor of local regions (K*K-pixel patch) in the image, can represent image more efficient than pixel-level intensity representation, and also than the popular Bag-Of-Feature (BOF) model, which approximately represents each local descriptor as a predefined visual word. Therefore it should be more effective in computational time than the BOF model. For extracting discriminant and compact features from the local descriptor tensor, we propose to use the proposed TMultilinear Manifold Analysis (MMA) algorithm, which has several benefits compared with conventional subspace learning methods such as PCA, ICA, LDA and so on: (1) a natural way of representing data without losing structure information, i.e., the information about the relative positions of pixels or regions; (2) a reduction in the small sample size problem which occurs in conventional supervised learning because the number of training samples is much less than the dimensionality of the feature space; (3) a neighborhood structure preserving in tensor feature space for face recognition and a good convergence property in training procedure. We validate our proposed algorithm on Benchmark database Yale and PIE, and experimental results show recognition rate with the proposed method can be greatly improved compared with conventional subspace analysis methods especially for small training sample number ....

      DOI: 10.1007/978-3-642-34166-3_81

      researchmap

    • Efficient Shape Representation and Statistical Shape Modeling of the Liver Using Spherical Harmonic Functions (SPHARM) 査読有り

      Tomoko Tateyama, Megumi Okegawa, Mei Uetani, Hidetoshi Tanaka, Shinya Kohara, Xianhua Han, Shuzo Kanasaki, Shigetaka Sato, Makoto Wakamiya, Akira Furukawa, Huiyan Jiang, Yen-Wei Chen

      6TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING AND INTELLIGENT SYSTEMS, AND THE 13TH INTERNATIONAL SYMPOSIUM ON ADVANCED INTELLIGENT SYSTEMS   428 - 431   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In the field of medical image analysis, the three-dimensional (3-D) shape representation and modeling of anatomic structures using only a few parameters is an important issue, and can be applied to computer assisted diagnosis, surgical simulations, visualization, and many other medical applications. In this paper, we show that the 3D anatomical structure such as the liver can be represented by a few coefficients of spherical harmonic functions (SPHARM). We also propose to use SPHARM based shape representation for statistical shape modeling. Since the dimension of SPHARM based shape representation vector is much lower than the conventional shape representation using coordinates of surface points, our proposed method can be used for small number of training samples and enhance the computation cost.

      DOI: 10.1109/SCIS-ISIS.2012.6505370

      researchmap

    • Super-Resolution of MR Volumetric Images Using Sparse Representation and Self-Similarity 査読有り

      Yutaro Iwamoto, Xian-Hua Han, So Sasatani, Kazuki Taniguchi, Wei Xiong, Yen-Wei Chen

      2012 21ST INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR 2012)   3758 - 3761   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Magnetic resonance imaging can only acquire volume data with finite resolution due to various factors. In particular, the resolution in the slice direction is much lower than that in the in-plane direction, yielding un-realistic visualizations. To solve this problem, interpolation techniques have conventionally been applied. However, classical interpolation techniques generally cause some artifact noise such as jaggedness and blurring in the edge regions. In this paper, we propose a new super-resolution framework for generating high-resolution data in the slice direction. In the proposed approach, we estimate the high-frequency component using a learning-based super-resolution technique with sparse representation and prove that the dictionary can be constructed using the in-plane frame as the input data without any other high-resolution data as training. Furthermore, we optimize estimated high-resolution data by adding a new regularization term with a non-local means algorithm. Experiments confirm that our proposed method is more effective than the conventional methods.

      researchmap

      その他リンク: http://dblp.uni-trier.de/db/conf/icpr/icpr2012.html#conf/icpr/IwamotoHSTXC12

    • Image Super-Resolution based on Locality-Constrained Linear Coding 査読有り

      Kazuki Taniguchi, Xian-Hua Han, Yutaro Iwamoto, So Sasatani, Yen-Wei Chen

      2012 21ST INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR 2012)   1948 - 1951   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper presents a learning-based method called image super-resolution (SR) for generating a high-resolution (HR) image from a single low-resolution (LR) image. Recent research investigated the image SR problem using sparse coding, which is based on good reconstruction of any image local patch by a sparse linear combination of atoms from an overcomplete dictionary. However, sparse-coding-based SR (ScSR) generally takes a significant amount of computational time to compute an HR image. Further, it can yield only a global dictionary D = [D-h;D-l] by jointly training the concatenated HR and LR image local patches, which results in no accurate correspondence between the HR and LR dictionaries. Therefore, we propose the generation of an HR image using a linear combination of several anchor points (codes) for a local patch based on locality-constrained linear coding (LLC), which is a fast implementation of local coordinate coding (LCC). In the proposed LLC-based strategy, each local patch is represented by a weighted linear combination of its nearer codes in a predefined codebook, and the linear weights become its local coordinate coding. Experimental results show that the recovered HR images with our proposed approach can achieve comparable performance at a processing time much shorter than those of conventional methods.

      researchmap

      その他リンク: http://dblp.uni-trier.de/db/conf/icpr/icpr2012.html#conf/icpr/TaniguchiHISC12

    • Group Sparse Representation of Adaptive Sub-Domain Selection for Image Classification 査読有り

      Xian-Hua Han, Xu Qiao, Yen-Wei Chen

      2012 21ST INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR 2012)   1431 - 1434   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Recent years have seen an increasing interest in codebook-based model(bag-of-words BOW) for image representation, which includes the basic bag-of-words model and its improved version for local descriptor reconstruction with sparse coding (SC) and locality-constrained linear coding (LLC) etc.. Although the recent coding strategies in the BoW model can lead to prospect performance using large amounts of codes (codebooks) for image classification, it is usually computational expensive for obtaining the global image representation through calculating the similarities between each local descriptor and all codes. Therefore, this study proposes to represent a local descriptor with an adaptive code or its variation modes (adaptive subdomain) in a small set of codebooks. The proposed strategy can adaptively select one code to saliently representation, or adaptively select one sub-domain of a code for group sparse reconstruction of a local descriptor. Due to computational cost mainly on the similarity calculation between local descriptors and the predefined codebooks, our proposed strategy using small set of codebook can greatly reduce computational time, and in addition, shows prospect performances for image classification on an scene database, called OMRON scene dataset, and the benchmark data: 15 natural scene dataset.

      researchmap

      その他リンク: http://dblp.uni-trier.de/db/conf/icpr/icpr2012.html#conf/icpr/HanQC12

    • Multiple Feature Selection and Fusion Based on Generalized N-Dimensional Independent Component Analysis 査読有り

      Danni Ai, Guifang Duan, Xianhua Han, Yen-Wei Chen

      2012 21ST INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR 2012)   971 - 974   2012年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper proposes a framework of tensor-based ICA method for N-dimensional data analysis, which is called generalized N-dimensional ICA (GND-ICA). The proposed GND-ICA is based on multilinear algebra that treats N-dimensional data as a tensor without any unfolding preprocess. As an application, the GND-ICA can be used for multiple feature fusion and representation for color image classification. Multiple features extracted from a given image are constructed as a tensor. The effective components for each feature can be selected simultaneously and combined by the GND-ICA. This can obtain the improved classification results in comparison with various conventional linear and multilinear subspace learning methods.

      researchmap

      その他リンク: http://dblp.uni-trier.de/db/conf/icpr/icpr2012.html#conf/icpr/AiDHC12

    • 勾配情報に基づくエッジ保存補間と超解像技術への応用

      岩本 祐太郎, 韓 先花, 健山 智, 大橋 基範, 笹谷 聡, 陳 延偉

      電気学会論文誌. C, 電子・情報・システム部門誌 = The transactions of the Institute of Electrical Engineers of Japan. C, A publication of Electronics, Information and Systems Society131 ( 11 ) 1901 - 1906   2011年11月1日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人 電気学会  

      Super-resolution is a process for obtaining high quality and high resolution images from a set of or only one low-resolution image. The most practical one for image super-resolution is reconstruction-based method, which minimizes the difference between observed low-resolution images and the estimation for high resolution image. Therein, interpolation step plays a key role for the estimated high resolution image quality. Usually, the conventional bilinear or bicubic methods are used in the reconstruction-based super-resolution. However, these conventional interpolations generally lead to blurring on edge regions and need more time for convergence in reconstruction-based super-resolution method. Therefore, this paper propose a gradient based edge preserving interpolation method, which can reduce not only artifact noise but also blurring near edge regions in the estimated high resolution image. Furthermore, our proposed interpolation method can also solve large complexity and time-consuming problem in the recently developed New Edge-directed interpolation, which usually can achieve sharp edge in the high resolution reconstructed image. Experiments validate that our proposed interpolation method for image super-resolution is more effective than the conventional interpolation ones.

      DOI: 10.1541/ieejeiss.131.1901

      Scopus

      CiNii Article

      researchmap

    • Global Selection vs Local Ordering of Color SIFT Independent Components for Object/Scene Classification 査読有り

      Dan-ni Ai, Xian-hua Han, Guifang Duan, Xiang Ruan, Yen-wei Chen

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE94D ( 9 ) 1800 - 1808   2011年9月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      This paper addresses the problem of ordering the color SIFT descriptors in the independent component analysis for image classification. Component ordering is of great importance for image classification, since it is the foundation of feature selection. To select distinctive and compact independent components (IC) of the color SIFT descriptors, we propose two ordering approaches based on local variation, named as the localization-based IC ordering and the sparseness-based IC ordering. We evaluate the performance of proposed methods, the conventional IC selection method (global variation based components selection) and original color SIFT descriptors on object and scene databases, and obtain the following two main results. First, the proposed methods are able to obtain acceptable classification results in comparison with original color SIFT descriptors. Second, the highest classification rate can be obtained by using the global selection method in the scene database, while the local ordering methods give the best performance for the object database.

      DOI: 10.1587/transinf.E94.D.1800

      J-GLOBAL

      researchmap

    • SCENE AND OBJECT RECOGNITION WITH SUPERVISED NONLINEAR NEIGHBORHOOD EMBEDDING 査読有り

      Xian-Hua Han, Yen-Wei Chen

      INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL7 ( 8 ) 4861 - 4870   2011年8月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:ICIC INT  

      Image category recognition is important to access visual information on the level of objects and scene types. In this paper, we develop a Supervised Nonlinear Neighborhood Embedding (SNNE) subspace algorithm of different visual features for object and scene recognition, which learns an adaptive nonlinear subspace by preserving the neighborhood structure of the visual feature space. In the proposed subspace algorithm, we combine the idea of nonlinear kernel mapping and preserving the neighborhood structure of the samples, so it can not only gain a perfect approximation of the nonlinear image manifold, but also enhance within-class neighborhood information. So, the proposed SNNE algorithm models the ensemble of visual features to a more discriminative space for category recognition, and at the same time, can effectively combine several visual features to improve recognition rate. The proposed method is evaluated by using the scene database (SIMPLicity) and object recognition database (Caltech). We confirm that the proposed method is much better than state-of-the-art methods only with simple visual features.

      researchmap

    • A robust method based on ICA and mixture sparsity for edge detection in medical images 査読有り

      Xian-Hua Han, Yen-Wei Chen

      SIGNAL IMAGE AND VIDEO PROCESSING5 ( 1 ) 39 - 47   2011年3月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:SPRINGER LONDON LTD  

      In this paper, a robust edge detection method based on independent component analysis (ICA) was proposed. It is known that most of the ICA basis functions extracted from images are sparse and similar to localized and oriented receptive fields. In this paper, the L (p) norm is used to estimate sparseness of the ICA basis functions, and then, the sparser basis functions were selected for representing the edge information of an image. In the proposed method, a test image is first transformed by ICA basis functions, and then, the high-frequency information can be extracted with the components of the selected sparse basis functions. Furthermore, by applying a shrinkage algorithm to filter out the components of noise in the ICA domain, we can readily obtain the sparse components of the noise-free image, resulting in a kind of robust edge detection even for a noisy image with a very low SN ratio. The efficiency of the proposed method for edge detection is demonstrated by experiments with some medical images.

      DOI: 10.1007/s11760-009-0140-5

      J-GLOBAL

      researchmap

    • 肝臓分割と血管抽出による肝臓情報の三次元可視化と手術支援システム (コンピュータビジョンとイメージメディア(CVIM) Vol.2011-CVIM-175)

      健山 智子, 進藤 典, Foruzan Amir Hossein

      情報処理学会研究報告2010 ( 5 ) 6p - 944   2011年2月

      詳細を見る

      記述言語:日本語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:情報処理学会  

      CiNii Article

      researchmap

    • Multi-class Co-training Learning for Object and Scene Recognition. 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      Proceedings of the IAPR Conference on Machine Vision Applications (IAPR MVA 2011), Nara Centennial Hall, Nara, Japan, June 13-15, 2011   67 - 70   2011年

    • CANONICAL CORRELATION ANALYSIS OF LOCAL FEATURE SET FOR VIEW-BASED OBJECT RECOGNITION 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      2011 18TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)   3601 - 3604   2011年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In this paper, we propose to use local feature set for image representation, which can represent variations in an object's appearance due to changing view point or camera pose. It was evidenced that usually only a part of the object are appeared in common when taking a photo of an object in different view points. With comparison of local features set extracted from different positions of images, an object can be recognized when common part is appeared in two images, which take photos of one object in different view points. In this paper, we use Canonical Correlation (also known as principle or canonical angles), which can be thought of as the angles between two d-dimensional subspace, as similarity measure of local feature sets. The proposed approach is evaluated in various view-based object datasets (Coil-100 and ETH80) for object and object category recognition. Experiments show that the performance advantages of our proposed approach can be achieved over existing techniques.

      DOI: 10.1109/ICIP.2011.6116496

      researchmap

    • HIGH FREQUENCY COMPENSATED FACE HALLUCINATION 査読有り

      So Sasatani, Xian-Hua Han, Takanori Igarashi, Motonori Ohashi, Yutaro Iwamoto, Yen-Wei Chen

      2011 18TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)   1529 - 1532   2011年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Face Hallucination is, one of a learning-based super-resolution technique that can reconstruct a high-resolution image using only one low-resolution image. However, there are often some detailed high-frequency components of the reconstructed image that cannot be recovered using this method. In this study, we proposed a high-frequency compensated face hallucination method for enhancing reconstruction performance. The proposed method can be divided into three steps: 1) high-resolution image reconstruction using a conventional hallucination method; 2) residual (high-frequency components) image recovery by "training" a residual image pair; 3) compensation of the reconstructed high-resolution image obtained in step 1 with the reconstructed residual image. Experimental results show that the high-resolution images obtained using our proposed approach are much better than those obtained by conventional hallucination.

      DOI: 10.1109/ICIP.2011.6115736

      researchmap

    • Independent component analysis of color SIFT for image classification 査読有り

      Dan-Ni Ai, Xian-Hua Han, Guifang Duan, Xiang Ruan, Yen-Wei Chen

      Proceedings - 2011 IEEE 7th International Conference on Intelligent Computer Communication and Processing, ICCP 2011   173 - 178   2011年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      This paper addresses the problems of feature selection and feature fusion. For the feature selection, the color SIFT descriptors in the independent components are ordered for image classification. To select distinctive and compact independent components (IC) of the color SIFT descriptors, we propose two ordering approaches based on variation: (1) Local ordering approaches (the localization-based ICs ordering and the sparseness-based ICs ordering) and (2) Global selection approach (PCA-based ICs selection).We evaluate the performance of proposed methods on object and scene databases, and obtain the following two main results. First, the proposed methods are able to obtain acceptable classification results in comparison with original color SIFT descriptors. Second, the highest classification rate can be obtained by using the global selection method in the scene database, while the local ordering methods give the best performance for the object database. For the aspect of feature fusion, tensor-based ICA is utilized to consider the relationship between different features. This obtains compact and distinctive representation of images for effective image classification. © 2011 IEEE.

      DOI: 10.1109/ICCP.2011.6047865

      Scopus

      researchmap

    • Multilinear Supervised Neighborhood Embedding with Local Descriptor Tensor for Face Recognition 査読有り

      Xian-Hua Han, Xu Qiao, Yen-Wei Chen

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE94D ( 1 ) 158 - 161   2011年1月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      Subspace learning based face recognition methods have attracted considerable interest in recent years, including Principal Component Analysis (PCA), Independent Component Analysis (ICA), Linear Discriminant Analysis (LDA). and some extensions for 2D analysis. However, a disadvantage of all these approaches is that they perform subspace analysis directly on the reshaped vector or matrix of pixel-level intensity, which is usually unstable under illumination or pose variance. In this paper, we propose to represent a face image as a local descriptor tensor, which is a combination of the descriptor of local regions (K*K-pixel patch) in the image, and is more efficient than the popular Bag-Of-Feature (BOF) model for local descriptor combination. Furthermore, we propose to use a multilinear subspace learning algorithm (Supervised Neighborhood Embedding-SNF) for discriminant feature extraction from the local descriptor tensor of face images, which can preserve local sample structure in feature space. We validate our proposed algorithm on Benchmark database Yale and PIE, and experimental results show recognition rate with our method can be greatly improved compared conventional subspace analysis methods especially for small training sample number.

      DOI: 10.1587/transinf.E94.D.158

      researchmap

    • Biomedical imaging modality classification using combined visual features and textual terms

      Xian-Hua Han, Yen-Wei Chen

      International Journal of Biomedical Imaging2011   1-7   2011年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)  

      We describe an approach for the automatic modality classification in medical image retrieval task of the 2010 CLEF cross-language image retrieval campaign (ImageCLEF). This paper is focused on the process of feature extraction from medical images and fuses the different extracted visual features and textual feature for modality classification. To extract visual features from the images, we used histogram descriptor of edge, gray, or color intensity and block-based variation as global features and SIFT histogram as local feature. For textual feature of image representation, the binary histogram of some predefined vocabulary words from image captions is used. Then, we combine the different features using normalized kernel functions for SVM classification. Furthermore, for some easy misclassified modality pairs such as CT and MR or PET and NM modalities, a local classifier is used for distinguishing samples in the pair modality to improve performance. The proposed strategy is evaluated with the provided modality dataset by ImageCLEF 2010. © 2011 Xian-Hua Han and Yen-Wei Chen.

      DOI: 10.1155/2011/241396

      Scopus

      researchmap

    • Color Independent Components Based SIFT Descriptors for Object/Scene Classification 査読有り

      Dan-ni Ai, Xian-hua Han, Xiang Ruan, Yen-wei Chen

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE93D ( 9 ) 2577 - 2586   2010年9月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      In this paper, we present a novel color independent components based SIFT descriptor (termed CIC-SIFT) for object/scene classification. We first learn an efficient color transformation matrix based on independent component analysis (ICA), which is adaptive to each category in a database. The ICA-based color transformation can enhance contrast between the objects and the background in an image. Then we compute CC-SIFT descriptors over all three transformed color independent components. Since the ICA-based color transformation can boost the objects and suppress the background, the proposed CIC-SIFT can extract more effective and discriminative local features for object/scene classification. The comparison is performed among seven SIFT descriptors, and the experimental classification results show that our proposed CC-SIFT is superior to other conventional SIFT descriptors.

      DOI: 10.1587/transinf.E93.D.2577

      researchmap

    • Color Independent Components Based SIFT Descriptors for Object/Scene Classification

      Dan-ni Ai, Xian-hua Han, Xiang Ruan, Yen-wei Chen

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE93D ( 9 ) 2577 - 2586   2010年9月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      In this paper, we present a novel color independent components based SIFT descriptor (termed CIC-SIFT) for object/scene classification. We first learn an efficient color transformation matrix based on independent component analysis (ICA), which is adaptive to each category in a database. The ICA-based color transformation can enhance contrast between the objects and the background in an image. Then we compute CC-SIFT descriptors over all three transformed color independent components. Since the ICA-based color transformation can boost the objects and suppress the background, the proposed CIC-SIFT can extract more effective and discriminative local features for object/scene classification. The comparison is performed among seven SIFT descriptors, and the experimental classification results show that our proposed CC-SIFT is superior to other conventional SIFT descriptors.

      DOI: 10.1587/transinf.E93.D.2577

      researchmap

    • Tensor-based subspace learning and its applications in multi-pose face synthesis 査読有り

      Xu Qiao, Xian-Hua Han, Takanori Igarashi, Keisuke Nakao, Yen-Wei Chen

      NEUROCOMPUTING73 ( 13-15 ) 2727 - 2736   2010年8月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:ELSEVIER SCIENCE BV  

      Facial pose synthesis is applied to generate much required information for several applications, such as public security, facial cosmetology, etc. How to synthesize facial pose images from one image accurately without spatial information is still a challenging problem. In this paper we propose a tensor-based subspace learning method (TSL) for synthesizing human multi-pose facial images from a single two-dimensional image. In the proposed TSL method, two-dimensional multi-pose images in the database are previously organized into a tensor form and a tensor decomposition technique is applied to build projection subspaces. In synthesis processing, the input two-dimensional image is first projected into its corresponding projection subspace to get an identity vector and then the identity vector is used to generate other novel pose images. Our technique is applied on KAO-Ritsumeikan Multi-angle View, Illumination and Cosmetic Facial Database (MaVIC) and experiment results show the effectiveness of our proposed method for facial pose synthesis. (C) 2010 Elsevier B.V. All rights reserved.

      DOI: 10.1016/j.neucom.2010.04.013

      J-GLOBAL

      researchmap

    • Independent Component Analysis-Based Prediction of O-Linked Glycosylation Sites in Protein Using Multi-Layered Neural Networks 査読有り

      Chu-Zheng Wang, Xiao-Feng Tan, Yen-Wei Chen, Xian-Hua Han, Masahiro Ito, Ikuko Nishikawa

      2010 IEEE 10TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS (ICSP2010), VOLS I-III   1761 - +   2010年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      In this paper, we develop a new method for prediction O-linked glycosylation site and pattern analysis in protein, which combines independent component analysis (ICA) with a multi-layer neural network (NN). ICA is first used to. construct main basis (subspace) of the protein sequence for features extraction. The projections of protein sequence on the subspace with low dimension are used as input data instead of the higher-dimensional protein sequences. Neural network is built to predict whether a particular site of serine or threonine is glycosylated. Compared with other subspace method, our proposed new method can improve the prediction accuracy.

      researchmap

    • RITSU_CBVR at TRECVID-2010. 査読有り

      Danni Ai, Atsushi Okamoto, Yae Kikutani, Yoshiyuki Tanaka, Xian-Hua Han, Yen-Wei Chen

      TRECVID 2010 workshop participants notebook papers, Gaithersburg, MD, USA, November 2010   2010年

      詳細を見る

      出版者・発行元:National Institute of Standards and Technology (NIST)  

      researchmap

      その他リンク: http://dblp.uni-trier.de/db/conf/trecvid/trecvid2010.html#conf/trecvid/AiAKYHC10

    • Multilinear Tensor Supervised Neighborhood Embedding Analysis for View-Based Object Recognition 査読有り

      Xian-Hua Han, Yen-Wei Cheni, Xiang Ruan

      ADVANCES IN MULTIMEDIA INFORMATION PROCESSING-PCM 2010, PT I6297   236 - 247   2010年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      In this paper, we propose a multilinear (N-Dimensional) Tensor Supervised Neighborhood Embedding (called ND-TSNE) for discriminant feature representation, which is used for view-based object recognition. ND-TSNE use a general Nth order tensor discriminant and neighborhood-embedding analysis approach for object representation. The benefits of ND-TSNE include: (1) a natural way of representing data without losing structure information, i.e., the information about the relative positions of pixels or regions; (2) a reduction in the small sample size problem which occurs in conventional supervised learning because the number of training samples is much less than the dimensionality of the feature space; (3) a neighborhood structure preserving in tensor feature space for object recognition and a good convergence property in training procedure. With Tensor-subspace features, the random forests as a multi-way classifier is used for object recognition, which is much easier for training and testing compared with multi-way SVM. We demonstrate the performance advantages of our proposed approach over existing techniques using experiments on the COIL-100 and the ETH-80 datasets.

      DOI: 10.1007/978-3-642-15702-8_22

      researchmap

    • Semi-supervised and interactive semantic concept learning for scene recognition 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      Proceedings - International Conference on Pattern Recognition   3045 - 3048   2010年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      In this paper, we present a novel semi-supervised and interactive concept learning algorithm for scene recognition by local semantic description. Our work is motivated by the continuing effort in content-based image retrieval to extract and to model the semantic content of images. The basic idea of the semantic modeling is to classify local image regions into semantic concept classes such as water, sunset, or sky [1]. However, labeling concept sampling manually for training semantic model is fairly expensive, and the labeling results is, to some extent, subjective to the operators. In this paper, by using the proposed semi-supervised and interactive learning algorithm, training samples and new concepts can be obtained accurately and efficiently. Through extensive experiments, we demonstrate that the image concept representation is well suited for modeling the semantic content of heterogenous scene categories, and thus for recognition and retrieval. Furthermore,higher recognition accuracy can be achieved by updating new training samples and concepts, which are obtained by the novel proposed algorithm. © 2010 IEEE.

      DOI: 10.1109/ICPR.2010.746

      Scopus

      researchmap

    • Image categorization by learned nonlinear subspace of combined visual-words and low-level features 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      Proceedings - International Conference on Pattern Recognition   3037 - 3040   2010年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      Image category recognition is important to access visual information on the level of objects and scene types. This paper presents a new algorithm for the automatic recognition of object and scene classes. Compact and yet discriminative visual-words and low-level-features object class subspaces are automatically learned from a set of training images by a Supervised Nonlinear Neighborhood Embedding (SNNE) algorithm, which can learn an adaptive nonlinear subspace by preserving the neighborhood structure of the visual feature space. The main contribution of this paper is two fold: i) an optimally compact and discriminative feature subspace is learned by the proposed SNNE algorithm for different feature space (visual-word and low-level features). ii) An effective merge of different feature subspace can be implemented simply. High classification accuracy is demonstrated on different database including the scene database (SIMPLicity) and object recognition database (Caltech). We confirm that the proposed strategy is much better than state-of-the-art methods for different databases. © 2010 IEEE.

      DOI: 10.1109/ICPR.2010.744

      Scopus

      researchmap

    • Adaptive color independent components based SIFT descriptors for image classification 査読有り

      Danni Ai, Xianhua Han, Xiang Ruan, Yen-Wei Chen

      Proceedings - International Conference on Pattern Recognition   2436 - 2439   2010年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      This paper proposes an adaptive color independent components based SIFT descriptor (termed CIC-SIFT) for image classification. Our motivation is to seek an adaptive and efficient color space for color SIFT feature extraction. Our work has two key contributions. First, based on independent component analysis (ICA), an adaptive and efficient color space is proposed for color image representation. Second, in this ICA-based color space, a discriminative CIC-SIFT descriptor is calculated for image classification. The experiment results indicate that (1) contrast between objects and background can be enhanced on the ICA-based color space and (2) the CIC-SIFT descriptor outperforms other conventional color SIFT descriptors on image classification. © 2010 IEEE.

      DOI: 10.1109/ICPR.2010.596

      Scopus

      researchmap

    • IMAGE RECOGNITION BY LEARNED LINEAR SUBSPACE OF COMBINED BAG-OF-FEATURES AND LOW-LEVEL FEATURES 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      2010 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING   1049 - 1052   2010年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Image category recognition is important to access visual information on the level of objects and scene types. This paper combines different feature representations of images and learn a compact subspace of different features for the automatic recognition of object and scene classes. Compact visual-words and low-level-features object class subspaces are automatically learned from a set of training images by a Regularized Linear Discriminant analysis (RLDA) algorithm, and the extracted RLDA-domain features are used for Support Vector Machine (SVM) classifier. The main contribution of this paper is two folds: i) Different features (bag-of-features and low-level features) is fused for image representation. ii) The compact feature subspaces (low-dimension features) of different features are learned for rendering to SVM classifier, which is computationally efficient for image category. High classification accuracy is demonstrated on object recognition database (Caltech). We confirm that the proposed strategy cam improve accuracy rate compared with state-of-the-art methods for object recognition databases.

      DOI: 10.1109/ICIP.2010.5653931

      researchmap

    • ImageCLEF 2010 Modality Classification in Medical Image Retrieval: Multiple Feature Fusion with Normalized Kernel Function. 査読有り

      Xian-Hua Han, Yen-Wei Chen

      CLEF 2010 LABs and Workshops, Notebook Papers, 22-23 September 2010, Padua, Italy   2010年

      詳細を見る

      出版者・発行元:CEUR-WS.org  

      researchmap

      その他リンク: http://dblp.uni-trier.de/db/conf/clef/clef2010w.html#conf/clef/Chen10

    • CONTRAST ENHANCEMENT OF MR BRAIN IMAGES BY CANONICAL CORRELATIONS BASED KERNEL INDEPENDENT COMPONENT ANALYSIS 査読有り

      Tomoko Tateyama, Zensho Nakao, Xianhua Han, Yen-Wei Chen

      INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL5 ( 7 ) 1857 - 1866   2009年7月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:ICIC INTERNATIONAL  

      Since the MR signals can be considered as a combination of the signals from each brain matters, it has been shown that independent component analysis (ICA) con be used for contrast enhancement of MR images. However, ICA is a linear method in nature, and it is inadequate to well-describe nonlinear variations of the real MR images. In this paper, we propose a new method for contrast enhancement of MR brain images using a canonical correlation based kernel independent component analysis (KICA). Experimental results on both phantom MR datasets and real clinical MR datasets show that the contrast of MR images can be significantly enhanced by KICA.

      researchmap

    • Contrarst Enhancement of MR Brain Images by Canonical Correlations Based Kernel Independent Component Analysis

      Tomoko Tateyama, Zensho Nakao, Xianhua Han, Yen-Wei Chen

      International Journal of Innovative Computing, Information and Control (IJICIC)5 ( 7 ) 1857-1866   2009年6月

      詳細を見る

    • Synthesis of multiple pose facial images using tensor-based subspace learning method 査読有り

      Xu Qiao, Yen-Wei Chen, Xian-Hua Han, Takanori Igarashi, Keisuke Nakao

      2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops 2009   219 - 226   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      Facial pose synthesis has many useful applications in practice. How to synthesize facial pose images robustly and simply is still a challenging problem. In this paper we proposed a tensor-based subspace learning method (TSL) that makes possible the synthesis of human multi-pose facial images from a single 2D image. We organize 2D multi-pose images in a tensor form and apply tensor decomposition to build a projection subspace. An input 2D image is projected into the projection subspace to get a corresponding identity vector. The identity vector is used to generate the novel pose images. The experiments are performed on MaVIC (KAO-Ritsumeikan Multi-angle View, Illumination and Cosmetic Facial Database) and preliminary experimental results show the effectiveness of our proposed method. ©2009 IEEE.

      DOI: 10.1109/ICCVW.2009.5457697

      Scopus

      researchmap

    • Supervised Local Subspace Learning for Region Segmentation and Categorization in High-Resolution Satellite Images 査読有り

      Yen-wei Chen, Xian-hua Han

      COMPUTATIONAL COLOR IMAGING5646   226 - 233   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      We proposed a new feature extraction method based on supervised locality preserving projections (SLPP) for region segmentation and categorization in high-resolution satellite images. Compared with other subspace methods such as PCA and ICA, SLPP can preserve local geometric structure of data and enhance within-class local information. The generalization of the proposed SLPP based method is discussed in this paper.

      DOI: 10.1007/978-3-642-03265-3_24

      researchmap

    • Image categorization by learned PCA subspace of combined visual-words and low-level features 査読有り

      Xian-Hua Han, Yen-Wei Chen

      IIH-MSP 2009 - 2009 5th International Conference on Intelligent Information Hiding and Multimedia Signal Processing   1282 - 1285   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      Image category recognition is important to access visual information on the level of objects and scene types. This paper combines different feature representations of images and learn a compact subspace of different features for the automatic recognition of object and scene classes. Compact visual-words and low-level-features object class subspaces are automatically learned from a set of training images by a Principle Component Analysis (PCA) algorithm, and the extracted PCA-domain features are used for Support Vector Machine (SVM) classifier. The main contribution of this paper is two fold: i) Different features (bag-of-features and low-level features)is fused for image representation. ii) The compact feature subspaces (low-dimension features) of different features are learned for rendering to SVM classifier, which is computationally efficient for image category. High classification accuracy is demonstrated on object recognition database (Caltech). We confirm that the proposed strategy is comparable with state-of-the-art methods for object recognition databases. © 2009 IEEE.

      DOI: 10.1109/IIH-MSP.2009.31

      Scopus

      researchmap

    • Principal component analysis for prediction of O-linked glycosylation sites in protein by multi-layered neural networks 査読有り

      Chu-Zheng Wang, Xian-Hua Han, Masahiro Ito, Ikuko Nishikawa, Yen-Wei Chen

      IIH-MSP 2009 - 2009 5th International Conference on Intelligent Information Hiding and Multimedia Signal Processing   1193 - 1196   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      Glycosylation is one of the common post-translation modification of protein in eukaryotic cells. Conventional neural network methods have been applied to predict glycosylation sites in protein sequence and the prediction accuracy is dependent on the dimension of feature vector (length of protein sequence). Though the prediction accuracy can be improved by increasing the length of protein sequence, it is time-consuming. In this paper, we propose a novel approach which combines PCA with a multilayer neural network for efficient, and accurate prediction of O-glycosylation sites in protein. PCA is first used to extract main basis (subspace) of the protein sequence. The lower-dimensional projections on the subspace are used as features instead of the higher-dimensional protein sequences. Compared with conventional method (without PCA), our proposed method can significantly reduce the larger computation cost, while can keep the higher prediction accuracy. © 2009 IEEE.

      DOI: 10.1109/IIH-MSP.2009.182

      Scopus

      researchmap

    • Image categorization with PCA-SICEF 査読有り

      Atsushi Okamoto, Xianhua Han, Xiang Ruan, Yen-Wei Chen

      5th International Conference on Natural Computation, ICNC 20096   31 - 35   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      Image category recognition is important to access visual information on the level of objects and scene types. This paper presents an automatic recognition system of scene and object with PCA-SICEF feature for digital color images. SICEF (Scale-Invariant Color and edge Feature) is an extension of the conventional local SIFT (Scale-Invariant Feature transform) feature, which only include edge invariance of local image region but not any color information. So the SIFT feature is not enough for distinguish image categorization especially for scene types, where the color information plays an important role for recognition. Therefore, we improve SIFT by including color feature for local image region, and name it as SICEF feature. However, the Dimension of the extracted SICEF feature is so high that we use PCA (Principle Component Analysis) to reduce the dimension, and then, use the PCA-domain SICEF (PCA-SICEF) for image classification. Experimental results show that it is much more efficient by our proposed PCA-SICEF feature than conventional SIFT feature. © 2009 IEEE.

      DOI: 10.1109/ICNC.2009.655

      Scopus

      researchmap

    • Hierarchical super-resolution approach for expanding image with high magnification 査読有り

      Motonori Ohashi, Xian-Hua Han, Yen-Wei Chen

      5th International Conference on Natural Computation, ICNC 20096   22 - 25   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE Computer Society  

      Recently, with the development of visual communication and image processing, there is a high demand for High-Resolution images such as video surveillance, medical imaging, and so on. Therefore, the Super-Resolution technology that produces a highresolution image from a set of shified, blurred, and decimated versions is actively researched. However, most previously published techniques perform well only for small magnifications but get worse either in computational complexity or ringing artifacts for large magnifications. In this paper, we propose a Hierarchical algorithm for high-magnification SuperResolution image reconstruction. The proposed algorithm magnifies the low-resolution image in multiple steps. Experiment results show that the new approach is more efficient and can provide much better reconstruction quality in comparison with the reconstruction result in one step. © 2009 IEEE.

      DOI: 10.1109/ICNC.2009.658

      Scopus

      researchmap

    • Object class recognition with supervised nonlinear neighborhood embedding of visual words 査読有り

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      1st International Conference on Internet Multimedia Computing and Service, ICIMCS 2009   25 - 28   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:ACM  

      This paper develops a supervised nonlinear subspace of bag-of-features for category classification. Bag-of-features represents an image as an orderless distribution of features, which selects the visual words by clustering and uses the similarity with each visual word as the features for classification. In this paper, we propose to model the ensemble of visual words with a supervised nonlinear neighborhood embedding method to a more discriminative space for category classification. The supervised nonlinear neighborhood embedding(SNNE) is used to model visual words and extract the discrimitive information specialized for each category. The projection length in subspace is used as features for classification. The SNNE subspace method can model the nonlinear variations induced by various kinds of visual words and extract more discriminative feature for object recognition. The proposed method is evaluated using the Cal-tech and GRAZ01 database. We confirm that the proposed method is comparable with state-of-the-art methods without absolute position information. Copyright 2009 ACM.

      DOI: 10.1145/1734605.1734615

      Scopus

      researchmap

    • Object class recognition with supervised nonlinear neighborhood embedding of visual words

      Xian-Hua Han, Yen-Wei Chen, Xiang Ruan

      1st International Conference on Internet Multimedia Computing and Service, ICIMCS 20097 ( 7 ) 25 - 28   2009年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      This paper develops a supervised nonlinear subspace of bag-of-features for category classification. Bag-of-features represents an image as an orderless distribution of features, which selects the visual words by clustering and uses the similarity with each visual word as the features for classification. In this paper, we propose to model the ensemble of visual words with a supervised nonlinear neighborhood embedding method to a more discriminative space for category classification. The supervised nonlinear neighborhood embedding(SNNE) is used to model visual words and extract the discrimitive information specialized for each category. The projection length in subspace is used as features for classification. The SNNE subspace method can model the nonlinear variations induced by various kinds of visual words and extract more discriminative feature for object recognition. The proposed method is evaluated using the Cal-tech and GRAZ01 database. We confirm that the proposed method is comparable with state-of-the-art methods without absolute position information. Copyright 2009 ACM.

      DOI: 10.1145/1734605.1734615

      Scopus

      researchmap

    • Edge detection algorithm based on ICA-domain shrinkage in noisy images

      Han XianHua, Dai ShuiYan, Li Jian, Xia GuoRong

      SCIENCE IN CHINA SERIES F-INFORMATION SCIENCES51 ( 9 ) 1349 - 1359   2008年9月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:SCIENCE PRESS  

      We propose a robust edge detection method based on ICA-domain shrinkage (independent component analysis). It is known that most basis functions extracted from natural images by ICA are sparse and similar to localized and oriented receptive fields, and in the proposed edge detection method, a target image is first transformed by ICA basis functions and then the edges are detected or reconstructed with sparse components. Furthermore, by applying a shrinkage algorithm to filter out the components of noise in ICA-domain, we can readily obtain the sparse components of the original image, resulting in a kind of robust edge detection even for a noisy image with a very low SN ratio. The efficiency of the proposed method is demonstrated by experiments with some natural images.

      DOI: 10.1007/s11432-008-0114-1

      researchmap

    • A spatio-chromatic ICA based noise reduction in color images 査読有り

      Xian-Hua Han, Yen-Wei Chen, Jian-Mei Le

      INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL4 ( 3 ) 661 - 669   2008年3月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:ICIC INT  

      In this paper, a spatio-chromatic ICA based noise reduction in color images is presented. We use RGB color images with a regular arrangement RGB channels into spatio-chromatic space, and ICA performs decorrelation of these signals. We show that spatio-chromatic components of images contain spatial information and color information, and that ICA spatio-chromatic analysis is able to help the reconstruction of images. We propose an efficient noise reduction method for color images based on the observation. In the proposed method, the image is first transformed to spatio-chromatic ICA domain and then the noise components are removed by soft thresholding (Shrinkage). Experimental results show that the denoised images based spatio-chromatic ICA domain shrinkage are dramatically improved in comparison to that with conventional filtering.

      researchmap

    • A SUPERVISED NONLINEAR NEIGHBORHOOD EMBEDDING OF COLOR HISTOGRAM FOR IMAGE INDEXING 査読有り

      Xian-Hua Han, Yen-Wei Chen, Takeshi Sukegawa

      2008 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, PROCEEDINGS   953 - +   2008年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Subspace learning. techniques are widespread in pattern recognition research. They include PCA, ICA, LPP, etc. These techniques are generally linear and unsupervised. The problem of image indexing is very complicated and the processed images are usually lie on non-linear image subspaces. In this paper, we propose. a supervised nonlinear neighborhood embedding: algorithm which learns an adaptive nonlinear subspace by preserving: the neighborhood structure of the image color.,pace. In the proposed algorithm, we combine the idea of nonlinear kernel mapping and preserving the neighborhood structure of the sampled, so it can not only gain a perfect approximation of the nonlinear image manifold, but also enhance within-class neighborhood information. Experimental results show that the proposed method outperform other linear or unsupervised subspace learning, methods.

      researchmap

    • Classification of high-resolution satellite images using supervised locality preserving projections 査読有り

      Yen-Wei Chen, Xian-Hua Han

      KNOWLEDGE-BASED INTELLIGENT INFORMATION AND ENGINEERING SYSTEMS, PT 2, PROCEEDINGS5178   149 - 156   2008年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      We proposed a new method based on Supervised locality preserving projections (SLPP) for classification of high resolution satellite images. Compared with other subspace methods Such as PCA and ICA, SLPP can preserve local geometric structure of data and enhance within-class local information. The proposed method has been successfully applied to IKONOS images and experimental results show that the proposed SLPP based method Outperform ICA-based method. The proposed method can be practically incorporated into a GIS system.

      DOI: 10.1007/978-3-540-85565-1-19

      DOI: 10.1007/978-3-540-85565-1_19

      researchmap

    • Enhancement and detection of lung nodules with multiscale filters in CT images 査読有り

      Shingo Takemura, Xianhua Han, Yen-Wei Chen, Kazuhiro Ito, Ikuko Nishikwa, Masahiro Ito

      2008 FOURTH INTERNATIONAL CONFERENCE ON INTELLIGENT INFORMATION HIDING AND MULTIMEDIA SIGNAL PROCESSING, PROCEEDINGS   717 - +   2008年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      It is important to enhance and detect nodules in CT images in order to identify the lung cancer at early stage. Nodules which are included in medical image generally have multiple size and scale, and have blob-like structure. Recently, 3D multiscale filter approach is proposed for lung nodules detection. However, the 3D method take too much computing time to be applicable in real diagnose systems. In this paper; we propose a two step method for lung nodules detection in CT images. Firstly, we use 2D multiscale filter to detect the candidates of lung nodules on the slice images, and then, reduce most of false positive nodules using logical AND operator of continuous CT slices. The effectiveness of the purposed methods has been demonstrated by Experiments.

      DOI: 10.1109/IIH-MSP.2008.303

      researchmap

    • Region-based segmentation and auto-annotation for color images 査読有り

      Yohei Tsurugai, Yuta Iwasaki, Xian-Hua Han, Yen-Wei Chen

      2008 FOURTH INTERNATIONAL CONFERENCE ON INTELLIGENT INFORMATION HIDING AND MULTIMEDIA SIGNAL PROCESSING, PROCEEDINGS   709 - 712   2008年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      This paper presents an auto-annotation system with simple pre-processed segmentation for digital color image. Recently, annotation techniques become one popular method for image retrieval system in image database management[1], image recognition system[2][3] and so on. In the paper, we propose a two-step approach for image annotation. Firstly, the color image is needed to be segmented into two parts: the main object is assumed as the foreground part, and the other will be the background part, and then, only the features of the foreground part (main object) are used for annotation of global images. Here, it is assumed that only a single main object is included in the color image, so annotation problem can be considered as a kind of classification problem, and all the images in database can be automatically categorized by neural network. After image annotating, user can easily retrieve the similar set of images with the same conception that user needs. Finally, we compare the performance of the proposed systems with different processing methods. Experimental results show efficiency of the proposed annotation system.

      DOI: 10.1109/IIH-MSP.2008.338

      researchmap

    • Robust face recognition based on modified ICA without training sample of test subjects 査読有り

      Xian-Hua Han, Yen-Wei Chen, Akihiko Yamada, Hideto Fujita

      2008 FOURTH INTERNATIONAL CONFERENCE ON INTELLIGENT INFORMATION HIDING AND MULTIMEDIA SIGNAL PROCESSING, PROCEEDINGS   701 - +   2008年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      At present there are many methods that could deal well with frontal view face recognition when there is sufficient number of representative training samples. Thereinto, sub-space learning method such as Principal component analysis(PCA), Independent component analysis(ICA), Linear Discriminant Analysis(LDA) are a very hot research topic in this field. However, in some face recognition system, the needed recognition faces are different with different users and are often changed according users' requirement. So in this paper, we proposed to make use of some known face database, in which subjects will be different with the test subjects, for training(extract subspace) with a modified ICA method. The proposed modified ICA method can save much of computer time and memory, and at the same time, can obtain acceptable experimental results on a part of FERET face database. In addition, we validate that the accuracy rate with simple rotation images of logged faces as known data can be improved when the logged face is only one sample per subject.

      DOI: 10.1109/IIH-MSP.2008.222

      researchmap

    • A SUPERVISED NONLINEAR NEIGHBORHOOD EMBEDDING OF COLOR HISTOGRAM FOR IMAGE INDEXING 査読有り

      Xian-Hua Han, Yen-Wei Chen, Takeshi Sukegawa

      2008 15TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1-5   949 - 952   2008年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Subspace learning techniques are widespread in pattern recognition research. They include PCA, ICA, LPP, etc. These techniques are generally linear and unsupervised. The problem of image indexing is very complicated and the processed images are usually lie on non-linear image subspaces. In this paper, we propose a supervised nonlinear neighborhood embedding algorithm which learns an adaptive nonlinear subspace by preserving the neighborhood structure of the image color space. In the proposed algorithm, we combine the idea of nonlinear kernel mapping and preserving the neighborhood structure of the samples, so it can not only gain a perfect approximation of the nonlinear image manifold, but also enhance within-class neighborhood information. Experimental results show that the proposed method outperform other linear or unsupervised subspace learning methods.

      DOI: 10.1109/ICIP.2008.4711913

      researchmap

    • Algorithm of ICA-based poisson-noise reduction and its application to CT imaging - art. no. 678908 査読有り

      Jian Li, Xianhua Han

      MIPPR 2007: MEDICAL IMAGING, PARALLEL PROCESSING OF IMAGES, AND OPTIMIZATION TECHNIQUES6789   78908 - 78908   2007年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      CT (computed tomography) imaging is a technology which uses X-ray beams (radiation) and computers to form detailed, cross-sectional images of an area of anatomy. However, the random scattered X-ray in CT imaging system will reduce radiographic contrast greatly in CT images. In this paper, a four-step method is proposed for decoding CT images: first, the EGSnrc Monte Carlo simulation system is used to simulate CT imaging and simulated data will be validated by real experimental data in the same experimental conditions; second, scattered X-ray image simulated by EGSnrc will be transformed into ICA-domain (independent component analysis-domain) to obtain the main magnitude of scattering data; third, a noise-reduction algorithm based on ICA-domain shrinkage is applied to smooth the CT image; fourth, the conventional linear deconvolution follows. The simulation results show that the reconstructed image is dramatically improved in comparison to that without the noise-removing filters, and the proposed method is also applied to real experimental X-ray imaging.

      researchmap

    • Independent component analysis for removing x-ray scatter in x-ray images 査読有り

      Yen-Wei Chen, Xianhua Han, Shiro Oikawa, Akinori Fujita

      2007 IEEE INSTRUMENTATION & MEASUREMENT TECHNOLOGY CONFERENCE, VOLS 1-5   2327 - +   2007年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Cone beam CT has a capability for the 3-dimentional imaging of large volumes with isotropic resolution. But the main limitation of cone beam CT is a larger amount of x-ray scatter. The x-ray scatter may enhance the noise in the reconstructed images. In this paper, we propose a new method based on independent component analysis (ICA) for removing the x-ray scatter in the observed x-ray images. The observed x-ray images can be considered as a linear mixture of scattered x-ray and primary x-ray. We show that the scattered x-ray and the primary x-ray can be blindly separated from two observed x-ray images by the use of ICA. The results indicate that the proposed method is effective for removing x-ray scatter from the x-ray images.

      researchmap

    • ICA-based noise reduction for PET sinogram-domain images 査読有り

      Xian-Hua Han, Yen-Wei Chen

      2007 IEEE/ICME INTERNATIONAL CONFERENCE ON COMPLEX MEDICAL ENGINEERING, VOLS 1-4   1655 - +   2007年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE  

      Projection data in Positron Emission Tomography (PET) are acquired as a number of photon counts from different observation angles. Positron decay is a random phenomenon that causes undesirably high variations in measured sinogram appearing as quantum noise. The ruduction of quantum noise or Poisson noise in medical images is an important issue. In this paper, we propose a new ICA-based filter for reduction of noise in sinogram domain. In the proposed filter, the sinogram (projection data) is firstly transformed to ICA domain, and then, the components of scattered projection are removed by a soft thresholding (Shrinkage). In this study, the choice of ICA basis function trained from different database is considered. The denoised results with different ICA basis fuinctions and conventional denoising method(wavelet shrinkage and Gaussian filter) are given for comparison, and then, we also show the reconstructed images of ICA-based denoised sinogram images using Filtered-Back-Projection(FBP) algorithm. Experimental results show that the reconstructed images of ICA-based denoised images are much clearer and have much better contrast than those without pre-processing filters.

      researchmap

    • Enhancement of IVR images by combining an ICA shrinkage filter with a multi-scale filter 査読有り

      Yen-Wei Chen, Kiyotaka Matsuo, Xianhua Han, Atsumoto Shimizu, Koichi Shibata, Yukio Mishina, Yoshihiro Mukuta

      REMOTE SENSING AND GIS DATA PROCESSING AND APPLICATIONS; AND INNOVATIVE MULTISPECTRAL TECHNOLOGY AND APPLICATIONS, PTS 1 AND 26790   2007年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      Interventional Radiology (IVR) is an important technique to visualize and diagnosis the vascular disease. In real medical application, a weak x-ray radiation source is used for imaging in order to reduce the radiation dose, resulting in a low contrast noisy image. It is important to develop a method to smooth out the noise while enhance the vascular structure. In this paper, we propose to combine an ICA Shrinkage filter with a multiscale filter for enhancement of IVR images. The ICA shrinkage filter is used for noise reduction and the multiscale filter is used for enhancement of vascular structure. Experimental results show that the quality of the image can be dramatically improved without any blurring in edge by the proposed method. Simultaneous noise reduction and vessel enhancement have been achieved.

      DOI: 10.1117/12.752036

      researchmap

    • Reconstruction of penumbral imaging based on a band-filtering algorithm 査読有り

      Xian-Hua Han, Chun Lin, ShuiYan Dai, Jian Li

      MIPPR 2007: MULTISPECTRAL IMAGE PROCESSING6787   2007年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      Penumbral imaging is a kind of technique which uses the facts that spatial information can be recovered from the shadow or penumbra that an unknown source casts through a simple large circular aperture. The technique is based on a linear deconvolution. As we known that the information of the penumbral image is only contained in the penumbra (the edges of the image), so according to that principle, we proposed a two-step method for decoding penumbral images in this paper. First, an edgy-emphasizing algorithm using a band filter is applied to extract the penumbras (the image edges) in noisy penumbral images; then, followed by conventional linear deconvolution of only the penumbral edges. The simulation results show that the reconstructed image is dramatically improved in comparison to that with the conventional noise-removing filters, and the proposed method is also applied to real experimental x-ray imaging.

      DOI: 10.1117/12.740386

      researchmap

    • An ICA based noise reduction for PET reconstructed images 査読有り

      Xian-Hua Han, Yen-Wei Chen, Keishi Kitamura, Akihiro Ishikawa, Yoshihiro Inoue, Kouichi Shibata, Yukio Mishina, Yoshihiro Mukuta

      2007 THIRD INTERNATIONAL CONFERENCE ON INTELLIGENT INFORMATION HIDING AND MULTIMEDIA SIGNAL PROCESSING, VOL 1, PROCEEDINGS   113 - +   2007年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      The reduction of noise in medical images is an important issue. In this paper we propose a new ICA-based filter for reduction of noise in reconstruction domain. In the proposed filter the reconstructed 3D PET images(X- Y plane-slice domain or X-Z plane) are firstly transformed to ICA domain, and then, the components of noise information are removed by a soft thresholding (Shrinkage). In this study, the choice of ICA basis function trained from noisy reconstructed images in different plane is considered. The denoised results with different ICA basis functions and conventional denoising method(wavelet shrinkage and Gaussian filter) are given for comparison. Experimental results show that the reconstructed images of ICA-based denoised images are much clearer and have much better contrast than those with wavelet-domain filters.

      DOI: 10.1109/IIH-MSP.2007.87

      researchmap

    • Application of Poisson image denoising by ICA to penumbral imaging 査読有り

      Xian-Hua Han, Han Li, ShuiYan Dai, Yen-Wei Chen

      FOURTH INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY, VOL 4, PROCEEDINGS   735 - +   2007年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:IEEE COMPUTER SOC  

      This paper proposes a new method based on independent component analysis (ICA) for Poisson noise reduction. In the proposed method, the image is first transformed to ICA domain and then the noise components are removed by a soft thresholding (Shrinkage). The proposed method, which is used as a preprocessing of the reconstruction, has been successfully applied to penumbral imaging. Both simulation results and experimental results show that the reconstructed image is dramatically improved in comparison to that without the noise-removing filters.

      DOI: 10.1109/FSKD.2007.180

      researchmap

    • An Edge Extraction Algorithm based ICA-Domain Shrinkage for Penumbral Imaging

      Xian-Hua Han, Zensho Nakao氏およびYen-Wei Chen氏

      INFORMATION9 ( 3 ) 529-544   2006年9月

      詳細を見る

    • On shrinkage-based edge detection in a PCA subspace for penumbral imaging 査読有り

      Xian-Hua Han, Zensho Nakao, Yen-Wei Chen

      INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL2 ( 1 ) 153 - 166   2006年2月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:ICIC INTERNATIONAL  

      We propose a robust edge extraction algorithm based on Principal component analysis (PCA) for Poisson noise reduction. In the proposed edge detection, the image is firstly transformed to PCA subspace with sparse PCA basis functions and then the noisy components are removed by a soft threshold (Shrinkage). The proposed edge extraction method, which is used as a preprocessing step of the reconstruction, has been successfully applied to penumbral imaging. Both simulation results and real experimental results show that the reconstructed images are dramatically improved in comparison to that with the conventional noise-removing filter.

      researchmap

    • ICA domain filtering for reduction of noise in x-ray images 査読有り

      Yen-Wei Chen, Xianhua Han

      Progress in Biomedical Optics and Imaging - Proceedings of SPIE6144   2006年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

      Radiological imaging such as x-ray CT is one of the most important tools for medical diagnostics. Since the radiological images are always with some quantum noise and the reduction of quantum noise or Poisson noise in medical images is an important issue. In this paper, we propose a new filtering based on independent component analysis (ICA) for reduction of noise. In the proposed filtering, the image (projection) is first transformed to ICA domain and then the components of scattered x-ray are removed by a soft thresholding (Shrinkage). The proposed method has been demonstrated by using both standard images and Monte Carlo simulations. Experimental results show that the quality of the image can be dramatically improved without any blurring in edge by the proposed filter.

      DOI: 10.1117/12.652043

      Scopus

      researchmap

    • ICA domain filtering for reduction of noise in x-ray images - art. no. 614469 査読有り

      Yen-Wei Chen, Xianhua Han

      Medical Imaging 2006: Image Processing, Pts 1-36144   14469 - 14469   2006年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      Radiological imaging such as x-ray CT is one of the most important tools for medical diagnostics. Since the radiological images are always with some quantum noise and the reduction of quantum noise or Poisson noise in medical images is an important issue. In this paper, we propose a new filtering based on independent component analysis (ICA) for reduction of noise. In the proposed filtering, the image (projection) is first transformed to ICA domain and then the components of scattered x-ray are removed by a soft thresholding (Shrinkage). The proposed method has been demonstrated by using both standard images and Monte Carlo simulations. Experimental results show that the quality of the image can be dramatically improved without any blurring in edge by the proposed filter.

      DOI: 10.1117/12.652043

      researchmap

    • An ICA-domain shrinkage based Poisson-noise reduction algorithm and its application to penumbral imaging

      XH Han, Z Nakao, YM Chen, R Kodama

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE88D ( 4 ) 750 - 757   2005年4月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      Penumbral imaging is a technique which exploits the fact that spatial information can be recovered from the shadow or penumbra that an unknown source casts through a simple large circular aperture. Since the technique is based on linear deconvolution, it is sensitive to noise. In this paper, a two-step method is proposed for decoding penumbral images: first, a noise-reduction algorithm based on ICA-domain (independent component analysis-domain) shrinkage is applied to smooth the given noise; second, the conventional linear deconvolution follows. The simulation results show that the reconstructed image is dramatically improved in comparison to that without the noise-removing filters, and the proposed method is successfully applied to real experimental X-ray imaging.

      DOI: 10.1093/ietsy/e88-d.4.750

      researchmap

    • An ICA-domain shrinkage based poisson-noise reduction algorithm and its application to penumbral imaging 査読有り

      Xian-Hua Han, Zensho Nakao, Yen-Wei Chen, Ryosuke Kodama

      IEICE Transactions on Information and SystemsE88-D ( 4 ) 750 - 757   2005年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:Institute of Electronics, Information and Communication, Engineers, IEICE  

      Penumbral imaging is a technique which exploits the fact that spatial information can be recovered from the shadow or penumbra that an unknown source casts through a simple large circular aperture. Since the technique is based on linear deconvolution, it is sensitive to noise. In this paper, a two-step method is proposed for decoding penumbral images: first, a noise-reduction algorithm based on ICA-domain (independent component analysis-domain) shrinkage is applied to smooth the given noise
      second, the conventional linear deconvolution follows. The simulation results show that the reconstructed image is dramatically improved in comparison to that without the noise-removing filters, and the proposed method is successfully applied to real experimental X-ray imaging. Copyright © 2005 The Institute of Electronics, Information and Communication Engineers.

      DOI: 10.1093/ietisy/e88-d.4.750

      Scopus

      researchmap

    • Speech Processing Algorithm with Combination of CIS Strategy for Cochlear Implants and Feature Extraction

      u Liu氏, Kai-Bao Nie氏, Xian-Hua HanおよびGuo-Xia Sun氏

      Chinese Journal of Biomedical Engineering25 ( 1 ) 70-73   2005年1月

      詳細を見る

    • Independent component analysis based filtering for penumbral imaging

      YW Chen, XH Han, S Nozaki

      REVIEW OF SCIENTIFIC INSTRUMENTS75 ( 10 ) 3977 - 3979   2004年10月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:AMER INST PHYSICS  

      We propose a filtering based on independent component analysis (ICA) for Poisson noise reduction. In the proposed filtering, the image is first transformed to ICA domain and then the noise components are removed by a soft thresholding (shrinkage). The proposed filter, which is used as a preprocessing of the reconstruction, has been successfully applied to penumbral imaging. Both simulation results and experimental results show that the reconstructed image is dramatically improved in comparison to that without the noise-removing filters. (C) 2004 American Institute of Physics.

      DOI: 10.1063/1.1787932

      researchmap

    • Robust edge detection by independent component analysis in noisy images

      XH Han, YW Chen, Z Nakao

      IEICE TRANSACTIONS ON INFORMATION AND SYSTEMSE87D ( 9 ) 2204 - 2211   2004年9月

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

      We propose a robust edge detection method based on independent component analysis (ICA). It is known that most of the basis functions extracted from natural images by ICA are sparse and similar to localized and oriented receptive fields, and in the proposed edge detection method, a target image is first transformed by ICA basis functions and then the edges are detected or reconstructed with sparse components only. Furthermore, by applying a shrinkage algorithm to filter out the components of noise in the ICA domain, we can readily obtain the sparse components of the original image, resulting in a kind of robust edge detection even for a noisy image with a very low SN ratio. The efficiency of the proposed method is demonstrated by experiments with some natural images.

      researchmap

    • Research Advance of implementing schemes of cochlear implant and its speech processing algorithms

      Xian-Hua Han, KaiBao Nie氏およびJu Liu氏

      Chinese Journal of Biomedical Engineering20 ( 2 ) 340-344   2003年2月

      詳細を見る

    • ICA-domain filtering of Poisson noise images 査読有り

      XH Han, YW Chen, Z Nakao, HQ Lu

      THIRD INTERNATIONAL SYMPOSIUM ON MULTISPECTRAL IMAGE PROCESSING AND PATTERN RECOGNITION, PTS 1 AND 25286   811 - 814   2003年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)   出版者・発行元:SPIE-INT SOC OPTICAL ENGINEERING  

      This paper proposes a new method to denoise images corrupted by Poisson noise. Poisson noise is signal-dependent, and consequently, separating signals from noise is a very difficult task. In most current Poisson noise reduction algorithms, noisy signal are pre-processed to approximate Gaussian noise, and then denoised by a conventional Gaussian denoising algorithm. In this paper, we propose to use adaptive basis functions derived from the data using modified ICA (Independent Component Analysis), and a maximum likelihood shrinkage algorithm based on the property of Poisson noise. This modified ICA method is based on a denoising method called "Sparse Code Shrinkage (SCS)" and wavelet-domain denoising. In denoising procedure of ICA-domain, the shrinkage function is determined by the property of Poisson noise that adapts to the intensity of signal. The performance of the proposed algorithm is validated with simulated data experiments, and the results demonstrate that the algorithm greatly improves the denoising performance in images contaminated by Poisson noise.

      researchmap

    • An ICA-based method for Poisson noise reduction 査読有り

      XH Han, YW Chen, Z Nakao

      KNOWLEDGE-BASED INTELLIGENT INFORMATION AND ENGINEERING SYSTEMS, PT 1, PROCEEDINGS2773   1449 - 1454   2003年

      詳細を見る

      記述言語:英語   掲載種別:研究論文(学術雑誌)   出版者・発行元:SPRINGER-VERLAG BERLIN  

      Many image systems rely on photon detection as a basis of image formation. One of the major sources of error in these systems is Poisson noise due to the quantum nature of the photon detection process. Unlike additive Gaussian noise, Poisson noise is signal dependent, and consequently separating signal from noise is a very difficult task. In most current Poisson noise reduction algorithms, noisy signal is firstly pre-processed to approximate Gaussian noise and then denoise by a conventional Gaussian denoising algorithm. In this paper, based on the property that Poisson noise adapts to the intensity of signal, we develop and analyze a new method using an optimal ICA-domain filter for Poisson noise removal. The performance of this algorithm is assessed with simulated data experiments and experimental results demonstrate that this algorithm greatly improves the performance in denoising image.

      DOI: 10.1007/978-3-540-45224-9_195

      researchmap

    • Speaker Recognition Using Combination of Fuzzy C-Mean(FCM) and Vector Quantification Algorithms

      Wu XiaoJuan氏, Xian-Hua HanおよびNie KaiBao氏

      Journal of electronics & Information Technology24 ( 6 ) 845-849   2002年6月

      詳細を見る

    ▼全件表示

    MISC

    • Attention機構を導入したCenterNetによる小麦の穂の検出

      井上源太, HAN Xian-Hua

      電子情報通信学会大会講演論文集(CD-ROM)2022   2022年

      詳細を見る

    • Robust Hyperspectral Image Super-Resolution via Constrained Non-negative Sparse Representation

      韓先花, 山口大, BoXin Shi(AIST, Yinqiang Zheng(NII, Toru Kouyama(AIST, Atsunori Kanemura(AIST, Ryosuke Nakamura(AIST

      第20回 画像の認識・理解シンポジウム   2017年8月

      詳細を見る

    • アルツハイマー病の画像診断―voxel-based morphometryと人工知能によるアルツハイマー病スコアの有用性

      椎野 顯彦, 岩本 祐太郎, 韓 先花, 陳 延偉

      脳循環代謝(日本脳循環代謝学会機関誌)28 ( 2 ) 303 - 308   2017年

      詳細を見る

      記述言語:日本語   出版者・発行元:日本脳循環代謝学会  

      <p>Voxel-based morphometry(VBM)はMRI画像から脳の局所的な体積を統計学的に解析するもので,z値として画像表記する手法はSPECTにおけるeZIS(easy z-score imaging system)やiSSP(interface software of 3D-SSP)と共通している.今回我々はVBM統合ソフトであるBAAD(Brain Anatomical Analysis using Diffeomorphic deformation)に人工知能(AI(artificial intelligence))を搭載し,複数の関心領域の情報からアルツハイマー病(Alzheimer's disease: AD)である確率をADS(Alzheimer's disease score)として提示させた.AIはRBF(radial basis function)カーネルによるSVM(support vector machine)を基本とし,スラック変数と境界マージン緩和の調整のための学習には北米のADNI(Alzheimer's disease neuroimaging initiative)データベース(AD=314,健常者=386)を用い,交差検証にはleave-one-out法を用いた.これによるADSの正答率と検査後オッズは89.6%,134.1であった.適合性の評価としてオーストラリアのAIBL(Australian imaging, biomarker & life style flagship study of ageing)データベース(AD=72,健常者=447)を用いVSRAD(voxel-based specific regional analysis system for Alzheimer's disease)と比較した.ADSの正答率と検査後オッズは86.1%と47.9で,VSRADの84.8%と14.9に比べて高い予測能力を示した.人工知能は多くの情報量を単純化することによって,日常診療における診断のサポートに役立つことが期待できる.本稿ではVBMの概要を紹介し,ADSの有用性についての検証結果を報告する.</p>

      DOI: 10.16977/cbfm.28.2_303

      CiNii Article

      researchmap

    • A NOVEL AND FAST CONNECTED COMPONENT-COUNTING ALGORITHM BASED ON GRAPH THEORY

      YANG SIHAI, CHEN DUANSHENG, HAN XIANHUA, CHEN YENWEI

      ICIC express letters. Part B, Applications : an international journal of research and surveys7 ( 12 ) 2625 - 2632   2016年12月

      詳細を見る

      記述言語:英語   出版者・発行元:ICIC International  

      CiNii Article

      researchmap

    • HEp-2 cell classification using K-support Spatial Pooling in Deep CNNs (医用画像)

      Han Xian-Hua, Chen Yen-Wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報116 ( 225 ) 1 - 6   2016年9月16日

      詳細を見る

      記述言語:英語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • HEp-2 cell classification using K-support Spatial Pooling in Deep CNNs (ヘルスケア・医療情報通信技術)

      Han Xian-Hua, Chen Yen-Wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報116 ( 224 ) 1 - 6   2016年9月16日

      詳細を見る

      記述言語:英語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • A preliminary study on tensor codebook model for multiphase medical image retrieval (ヘルスケア・医療情報通信技術)

      WANG Jian, Han Xian-hua, Xu Yingying, Lin Lanfen, Hu Hongjie, Jin Chongwu, Chen Yen-Wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報116 ( 224 ) 47 - 50   2016年9月16日

      詳細を見る

      記述言語:英語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • A-2-3 Liver lesion retrieval based on multiphase medical volumes using sparse coding based codebook model

      WANG Jian, Han Xian-hua, Xu Yingying, Lin Lanfen, Hu Hongje, Jin Chongwu, Chen Yen-Wei

      電子情報通信学会基礎・境界ソサイエティ/NOLTAソサイエティ大会講演論文集2016   33 - 33   2016年3月1日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • D-12-16 WLTPを用いた食事画像認識(D-12.パターン認識・メディア理解A(パターンメディアの認識・理解・生成),一般セッション)

      藤田 和哉, 笹野 翔太, 韓 先花, 陳 延偉

      電子情報通信学会総合大会講演論文集2016 ( 2 ) 85 - 85   2016年3月1日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • D-12-17 データ駆動型モデルを用いた食事画像認識(D-12.パターン認識・メディア理解A(パターンメディアの認識・理解・生成),一般セッション)

      笹野 翔太, 韓 先花, 陳 延偉

      電子情報通信学会総合大会講演論文集2016 ( 2 ) 86 - 86   2016年3月1日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • 複数ボリュームデータを用いたマウスMR画像の超解像処理 (画像工学)

      岩本 祐太郎, 韓 先花, 椎野 顯彦

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報115 ( 459 ) 35 - 39   2016年2月22日

      詳細を見る

      記述言語:日本語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • 深層学習による医用画像の高解像度化 (マルチメディアストレージ コンシューマエレクトロニクス ヒューマンインフォメーション メディア工学 映像表現&コンピュータグラフィックス)

      近藤 佑斗, 趙 キョウ楚, 岩本 祐太郎, 韓 先花, 陳 延偉

      映像情報メディア学会技術報告 = ITE technical report40 ( 6 ) 29 - 34   2016年2月

      詳細を見る

      記述言語:日本語   出版者・発行元:映像情報メディア学会  

      CiNii Article

      researchmap

    • 複数ボリュームデータを用いたマウスMR画像の超解像処理 (マルチメディアストレージ コンシューマエレクトロニクス ヒューマンインフォメーション メディア工学 映像表現&コンピュータグラフィックス)

      岩本 祐太郎, 韓 先花, 椎野 顯彦, 陳 延偉

      映像情報メディア学会技術報告 = ITE technical report40 ( 6 ) 35 - 39   2016年2月

      詳細を見る

      記述言語:日本語   出版者・発行元:映像情報メディア学会  

      CiNii Article

      researchmap

    • Content-based retrieval of focal liver lesions using sparse representations of multiphase contrast-enhanced CT images (医用画像)

      WANG Jian, Han Xian-hua, Xu Yingying

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報115 ( 401 ) 171 - 176   2016年1月19日

      詳細を見る

      記述言語:英語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • Deep Convolutional Neural Networkによる食事画像認識 (パターン認識・メディア理解)

      韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報115 ( 224 ) 67 - 72   2015年9月14日

      詳細を見る

      記述言語:日本語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • 生活支援 スパース表現とハイブリッドプーリングを用いた高精度な食事画像認識

      韓 先花, 楠本 理子, 陳 延偉

      画像ラボ26 ( 6 ) 1 - 7   2015年6月

      詳細を見る

      記述言語:日本語   出版者・発行元:日本工業出版  

      CiNii Article

      researchmap

    • Sparse and Low Rank DecompositionとICPを用いた頑健な点群位置合わせ法 (信号処理)

      ちょう きょうそ, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報115 ( 22 ) 23 - 28   2015年5月14日

      詳細を見る

      記述言語:日本語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • 正規化スパース性制約を用いたぼけカーネル推定および動画像復元への応用 (信号処理)

      野島 優補, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報115 ( 22 ) 77 - 82   2015年5月14日

      詳細を見る

      記述言語:日本語   出版者・発行元:電子情報通信学会  

      CiNii Article

      researchmap

    • 認識・検出 データ駆動型モデルを用いた画像認識

      韓 先花, 中島 基輝, 陳 延偉

      画像ラボ26 ( 5 ) 14 - 19   2015年5月

      詳細を見る

      記述言語:日本語   出版者・発行元:日本工業出版  

      CiNii Article

      researchmap

    • D-11-34 2段階Hallucination法による医用画像の高解像度化と3次元医用画像への応用(D-11.画像工学,一般セッション)

      近藤 佑斗, 韓 先花, 陳 延偉

      電子情報通信学会総合大会講演論文集2015 ( 2 ) 34 - 34   2015年2月24日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • D-12-19 3次元顔面形態の局所的統計解析(D-12.パターン認識・メディア理解A(パターンメディアの認識・理解・生成),一般セッション)

      中津 美冴, 韓 先花, 木村 亮介, 陳 延偉

      電子情報通信学会総合大会講演論文集2015 ( 2 ) 71 - 71   2015年2月24日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • D-16-9 ベイズモデルを用いたCT画像からの腫瘍候補の検出(D-16.医用画像,一般セッション)

      今野 悠, 韓 先花, Wei Xiong, 陳 延偉

      電子情報通信学会総合大会講演論文集2015 ( 2 ) 149 - 149   2015年2月24日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • B-20-9 PLS回帰を用いた3次元顔面形態解析(B-20.ヘルスケア・医療情報通信技術,一般セッション)

      中津 美冴, 韓 先花, 木村 亮介, 陳 延偉

      電子情報通信学会ソサイエティ大会講演論文集2014 ( 1 ) 421 - 421   2014年9月9日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • B-20-11 PLS回帰を用いた肝硬変症支援診断(B-20.ヘルスケア・医療情報通信技術,一般セッション)

      上谷 芽衣, 健山 智子, 小原 伸哉, 韓 先花, 金崎 周造, 井上 明星, 古川 顕, 陳 延偉

      電子情報通信学会ソサイエティ大会講演論文集2014 ( 1 ) 423 - 423   2014年9月9日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • Sparse and Low Rank Matrix Decomposition for Cirrhosis Diagnosis based Local Morphological Analysis (医用画像)

      Deng Junping, Han Xian-Hua, Chen Yen-Wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 200 ) 33 - 37   2014年9月2日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Cirrhosis liver is a terrible disease which is threatening our lives. Meanwhile, cirrhosis will cause significant hepatic morphological changes. While it is well known that the livers from different subjects have similar global shape structure which means liver shape ensemble should be low-rank. However the deformation which caused by cirrhosis can be considered as sparse compared with the whole liver. Therefore, in this study, we proposed to apply spare and low-rank matrix decomposition to partition the local deformation part (sparse error matrix E) from the global similar structure (low-rank matrix A) using the input liver shape D, which is the landmark coordinates of liver shapes and already have been aligned by the current rigid registration methods firstly. And then sparse matrix E is used for diagnosis. In common sense, the normal liver should have less local deformation than that of abnormal liver, which means that the norm of sparse matrix E for normal liver is smaller than the norm for abnormal one. Thus, we proposed a method which found a threshold classifier to classify normal and abnormal livers using the norm of E for these two categories. The proposed method is evaluated by a liver database and compared with statistical shape model(SSM) based methods. The experimental results of proposed method is better than those of SSM-based methods.

      CiNii Article

      researchmap

    • Stacked Fisher Networkを用いたHEp-2細胞認識 (医用画像)

      韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 200 ) 21 - 26   2014年9月2日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      本研究ではHEp-2細胞の自動認識を目的する.HEp-2細胞認識において,如何に細胞画像からコンパクト且つ有効な特徴を抽出し画像表現するのは認識精度に大きく影響される.本研究では,画像表現を注目し,より判別的な特徴を抽出ために以下の工夫を行う:(1)人間の認知理論に基づいて画像のrawドメインからexcitationドメインへの変更;(2)量子化ではなく直接局所的なパッチ情報でマイクロの使用;(3)混合ガウスモデルを用いたfisher networkの構築;(4)構築したfisher networkを積み重ねることで深層フレームワーク(Stacked Fisher Network)への発展ICIP2013で開催されたHEp-2細胞認識コンテストのデータベースを用いた実験による,従来手法に比べで提案したフレームワークの認識精度向上を検証した.

      CiNii Article

      researchmap

    • Improved Interactive Medical Image Segmentation using Graph Cut and Superpixels (Medical Imaging)

      Kitrungrotsakul Titinunt, Dong Chunhua, Han Xian-Hua, Chen Yen-Wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 103 ) 17 - 20   2014年6月24日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Interactive image segmentation is a useful method for selecting object of interest in image. The variations of intensity and shape in medical images (organs) limits their ability to precisely localize object boundaries, computation time of segmentation and therefore lack of accuracy in the segmentation object. The popular interactive segmentation method is Graph Cut. The computation time of each cut is a key to make interactive image segmentation useful in real application usage. The generally of medical images are larger than 2D image. The lack of computation time will be occur if we try to apply segment out the object in the medical images using only Graph Cut. This paper presents a method for combining Graph Cut with SLIC (Simple Linear Iterative Clustering) to adapt to medical image. To be precise, our method is initialized by design superpixels with SLIC super pixels. With SLIC superpixels, we can increasing the accuracy and also boost up computation time of Graph Cut. The experiments show segmentation results with our method is significantly better than only using Graph Cut in term of accuracy and computation time.

      CiNii Article

      researchmap

    • Organ Bounding Box Annotation based on Adaptive Selection of Bone References (Medical Imaging)

      DONG Chunhua, FORUZAN Amir H., HAN Xian-hua, TATEYAMA Tomoko, CHEN Yen-wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 103 ) 27 - 32   2014年6月24日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Accurate segmentation of abdominal organs is a key step in computer-aided diagnosis (CAD) system. To accurately segment a tissue, a volume of interest (VOI), in which the organ was located, was defined by a user. However, user interaction usually makes the task laborious and time-consuming. Hence, we propose a method that finds the VOI bounding box of the organ automatically based on the bone reference, which can be easily extracted from CT volumes. The basic idea of our proposed method is to achieve anatomical localization according to the statistical geometric location of organs within the bone reference. How to choose the bone as the reference was a difficult task, because the available abdominal volume have large variation during imaging procedure. With taking these into consideration, we prepared four different bone references. Using the adaptive selection of appropriate reference bone, the extracted bone from the input image is registered to the reference. After registration of all tissues of the training images, we find the VOI of the ensemble of tissues and use it as the organ bounding box. For the test images, according to the adaptive selection of bone reference, the candidate organ region is extracted based on this organ bounding box. We demonstrated the effectiveness of our method by finding the bounding box of kidney, liver and spleen.

      CiNii Article

      researchmap

    • Multi-touch Based Medical Interactive Visualization System (パターン認識・メディア理解)

      Wang Jian, Tu Hua-Wei, Han Xian-Hua, Tateyama Tomoko, Chen Yen-Wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 41 ) 31 - 35   2014年5月22日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Medical imaging plays a central role in many healthcare practices. With the development of medical imaging devices, the medical image data can be achieved in higher and higher definitions, and then how to analyze and visualize the acquired large-amount and complex data is a desiring demand in medical education and clinical fields. Therefore, this paper develops a multi-touch based medical image analysis and visualization system for analyzing medical images, displaying the whole body's organs of a subject and the interested organ and region to explore more detailed structures. The designed system includes three modules: (1) Image analysis module, which will enable user segment a desired organ interactively; (2) Visualization module, which can manifest not only the detailed information and global structures of an organ in different view-points but also the correlated location relations between different organs; (3) Control module, which can enable the user to easily interact with the system by multi-touch based control. The designed system is able to achieve a detail visualization of the integrated components, and can be applied to education, doctor-training and clinical sites.

      CiNii Article

      researchmap

    • Multi-touch Based Medical Interactive Visualization System (信号処理)

      Wang Jian, Tu Hua-Wei, Han Xian-Hua, Tateyama Tomoko, Chen Yen-Wei

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 39 ) 31 - 35   2014年5月22日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Medical imaging plays a central role in many healthcare practices. With the development of medical imaging devices, the medical image data can be achieved in higher and higher definitions, and then how to analyze and visualize the acquired large-amount and complex data is a desiring demand in medical education and clinical fields. Therefore, this paper develops a multi-touch based medical image analysis and visualization system for analyzing medical images, displaying the whole body's organs of a subject and the interested organ and region to explore more detailed structures. The designed system includes three modules: (1) Image analysis module, which will enable user segment a desired organ interactively; (2) Visualization module, which can manifest not only the detailed information and global structures of an organ in different view-points but also the correlated location relations between different organs; (3) Control module, which can enable the user to easily interact with the system by multi-touch based control. The designed system is able to achieve a detail visualization of the integrated components, and can be applied to education, doctor-training and clinical sites.

      CiNii Article

      researchmap

    • スパースコーディングとハイブリッドプーリングを用いた食事画像認識 (医用画像)

      楠元 理子, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 42 ) 1 - 5   2014年5月22日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      近年,食事の乱れや健康意識の向上に伴い,食生活の管理が重要となっている.そこで,携帯電話のカメラ機能と用いて撮影された食事画像から自動で食事内容の認識を行なうことで,健康を容易に促進できるシステムの構築を目指している.近年では画像特徴であるBag of Featuresを用いた画像認識で,ある程度の認識精度を得ることができるようになった.しかし,Bag of Featuresではベクトル量子化を用いて局所特徴を表現するが,量子化の時点で多くの情報が欠落する.本研究では食事画像の認識精度を改良するために,sparse codingを用いてより少ない情報の欠落で局所特徴を表現して,画像認識を行う.さらに,従来のプーリング法を改善し,新たなTop-Ranked Average-Poolingを提案した.本研究で提案した手法は2つの食事データベースにおいて,高い認識精度が得られることを検証した.

      CiNii Article

      researchmap

    • 2段階Hallucination法および3次元医用画像の高解像度化への応用 (医用画像)

      近藤 佑斗, 野島 優補, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 42 ) 61 - 66   2014年5月22日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      医療の現場では高解像度の画像が必要不可欠であるが, MRにより高解像度の画像を得るためには長い撮像時間が必要となり患者に多くの負担を強いることになる.そこで,低解像度の画像から高解像度の画像を生成することの可能な超解像技術に注目が集まっている.本研究では,単一の低解像度画像から高解像度画像を生成するSingle-Frame SRの手法の中で,データベースから複数のサンプルを選び再構成することで大規模なデータベースを用意することなく精度の高い高解像度化が可能となるImage Hallucination法を用いた.しかし,従来のImage Hallucination法では十分に高周波成分を再現できないという問題点があった.そこで,本研究では2段階Hallucination法を用いた高周波成分を高精度に再現できる超解像技術を提案した.また,本手法を3次元画像にも適応し,解像度の高いAxial平面の画像データベースを用いることでスライス間隔に依存して解像度に制限のかかる他平面画像の高解像度化が可能であることを証明した.

      CiNii Article

      researchmap

    • 統計学習法を用いた3次元顔面形態解析 (医用画像)

      中津 美冴, 韓 先花, 木村 亮介, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 42 ) 67 - 72   2014年5月22日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      遺伝子に関係した研究が様々な分野で行われており,顔面形態との関係性も注目されている.本研究では先行研究で提案された3次元顔面形態解析のフレームワークに基づく従来の複雑な非剛体対応付け方法に対し,円筒座標変換を用いた簡単かつ効率的な対応付け法を提案した.また,顔面形態を表現するため主成分に新たなMean Hyperplaneの特徴を加え,各特徴を正規化することで日本列島と沖縄地方の地域分類の精度向上を検証した.

      CiNii Article

      researchmap

    • ノンパラメトリック確率モデルを用いたCT画像からの腫瘍候補の検出 (医用画像)

      今野 悠, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 42 ) 73 - 78   2014年5月22日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      本研究ではCT肝臓画像から腫瘍候補を自動的に検出する事を目的とする.従来法ではCT画像は肝臓,血管と腫瘍を含むガウス混合分布と仮定し, EM法等を用いてパラメータの導出及び腫瘍の有無の分類を行う.しかし,従来法では入力CTボリュームデータのすべでのボクセルを用いてパラメータのを推定し,計算コストが高いだけではなく,モデリング結果から腫瘍モデルの選別する際に手作業での分類が必要である。また,腫瘍部分が一つのガウス分布として存在すると仮定しているため腫瘍なしのデータに対し正しくモデリングできないという問題点がある。そこで本研究ではCT画像から簡単に切り出しできる肝臓と血管のプロトタイプを用意し,ノンパラメトリック確率モデルを用いてCT画像から腫瘍候補の検出を提案する.

      CiNii Article

      researchmap

    • Generalized Super-Vectorを用いた一般画像分類 (医用画像)

      中島 基輝, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告 = IEICE technical report : 信学技報114 ( 42 ) 113 - 117   2014年5月22日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      近年,計算機を用いた画像認識技術は進歩しており,認識率の大幅な改善を図った手法が提案されている.その1つとして,局所特徴を用いたBag-Of-features(BOF)が挙げられる.しかし,これは計算が複雑なSIFT特徴を用いており,計算コストが大きいという問題がある.そこで,色や形状のヒストグラムを表す計算コストが小さいRaw特徴を用いた新しい特徴抽出のフレームワークを提案する.また,特徴記述の際により高次な情報を加えたSuper-Vector Cordingを用いて,一般画像認識の精度向上の可能性を検討する.そこで,抽出した特徴についてSVMによる学習と分類の実験を行い,従来のBOFと比べて高速かつ高精度な分類を示すことができた.

      CiNii Article

      researchmap

    • Sparse Dictionary Representation and Propagation for MRI Volume Super-Resolution (パターン認識・メディア理解)

      Han Xian-Hua, Chen Yen-Wei

      電子情報通信学会技術研究報告. PRMU, パターン認識・メディア理解112 ( 37 ) 13 - 18   2012年5月10日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      This study addresses the problem of generating a high-resolution (HR) MRI volume from a single low-resolution (LR) MRI input volume. Recent researches have proved that sparse coding can be successfully applied for single-frame super-resolution for natural images, which is based on good reconstruction of any local image patch with a sparse linear combination of atoms taken from an appropriate over-complete dictionary. This study adapts the basic idea of sparse code-based super-resolution (SCSR) for MRI volume data, and then improves the dictionary learning strategy in the conventional SCSR for achieving the precise sparse representation of HR volume patches. In the proposed MRI super-resolution strategy, we only learn the dictionary of the HR MRI volume patches with sparse coding algorithm, and then propagate the HR dictionary to the LR dictionary by mathematical analysis for calculating the sparse representation (coefficients) of any LR local input volume patch. The unknown corresponding HR volume patch can be reconstructed with the sparse coefficients from the LR volume patch and the corresponding HR dictionary. We validate that the proposed SCSR strategy through dictionary propagation can recover much clearer and more accurate HR MRI volume than the conventional interpolated methods.

      CiNii Article

      researchmap

    • Generalized N-Dimensional Independent Component Analysis Based Multiple Feature Selection and Fusion (パターン認識・メディア理解)

      Ai Danni, Duan Guifang, Han Xianhua, CHEN Yen-Wei

      電子情報通信学会技術研究報告. PRMU, パターン認識・メディア理解112 ( 37 ) 81 - 86   2012年5月10日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      We proposed a multilinear independent component analysis framework called generalized N-dimensional ICA (GND-ICA) by extending the conventional linear ICA based on multilinear algebra. Unlike the linear ICA that only treats one-dimensional data, the proposed GND-ICA treats N-dimensional data as a tensor without any preprocess of data vectorization. We furthermore introduce two types of GND-ICA solutions and analysis their efficiency and effectiveness. As an application, the GND-ICA can be used for multiple feature fusion and representation for color image classification. Many features extracted from a given image are constructed as a tensor. The feature tensor can be effective represented by GND-ICA. Compared with conventional linear subspace learning methods, GND-ICA is capable of obtaining more distinctive representation for color image classification.

      CiNii Article

      researchmap

    • A Novel Components Selection Method in Statistical Shape Model for Computer-aided Diagnosis : And it's Application for Assisting Cirrhosis Diagnosis (パターン認識・メディア理解)

      Luo Jie, Han Xian-Hua, Kohara Shinya, TATEYAMA Tomoko, FURUKAWA Akira, KANAZAKI Shuzo, CHEN Yen-Wei

      電子情報通信学会技術研究報告. PRMU, パターン認識・メディア理解112 ( 37 ) 33 - 38   2012年5月10日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      In the field of Computer-aided Diagnosis, organs' shape variability have been found to be important for understanding their inherent structures. It's also well known that Statistical Shape Model (SSM) can effectively capture the shape variation of given training set. Consequently, diseases that cause organ deformations have the possibility to be diagnosed with the aid of computer by SSM based organ shape analysis. However, it's difficult to decide which modes are selected as indicators to assist the diagnosis. Hence, applications of applying SSM for assisting diagnosis haven't widely concerned. In this study we propose a novel modes selection method and test it in case of assisting Cirrhosis diagnosis. Experimental results validate the effectiveness of our method and its potential for assisting the clinic diagnosis.

      CiNii Article

      researchmap

    • A Novel Components Selection Method in Statistical Shape Model for Computer-aided Diagnosis : And it's Application for Assisting Cirrhosis Diagnosis (画像工学)

      Luo Jie, Han Xian-Hua, Kohara Shinya, TATEYAMA Tomoko, FURUKAWA Akira, KANAZAKI Shuzo, CHEN Yen-Wei

      電子情報通信学会技術研究報告. IE, 画像工学112 ( 36 ) 33 - 38   2012年5月10日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      In the field of Computer-aided Diagnosis, organs' shape variability have been found to be important for understanding their inherent structures. It's also well known that Statistical Shape Model (SSM) can effectively capture the shape variation of given training set. Consequently, diseases that cause organ deformations have the possibility to be diagnosed with the aid of computer by SSM based organ shape analysis. However, it's difficult to decide which modes are selected as indicators to assist the diagnosis. Hence, applications of applying SSM for assisting diagnosis haven't widely concerned. In this study we propose a novel modes selection method and test it in case of assisting Cirrhosis diagnosis. Experimental results validate the effectiveness of our method and its potential for assisting the clinic diagnosis.

      CiNii Article

      researchmap

    • Generalized N-Dimensional Independent Component Analysis Based Multiple Feature Selection and Fusion (画像工学)

      Ai Danni, Duan Guifang, Han Xianhua, CHEN Yen-Wei

      電子情報通信学会技術研究報告. IE, 画像工学112 ( 36 ) 81 - 86   2012年5月10日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      We proposed a multilinear independent component analysis framework called generalized N-dimensional ICA (GND-ICA) by extending the conventional linear ICA based on multilinear algebra. Unlike the linear ICA that only treats one-dimensional data, the proposed GND-ICA treats N-dimensional data as a tensor without any preprocess of data vectorization. We furthermore introduce two types of GND-ICA solutions and analysis their efficiency and effectiveness. As an application, the GND-ICA can be used for multiple feature fusion and representation for color image classification. Many features extracted from a given image are constructed as a tensor. The feature tensor can be effective represented by GND-ICA. Compared with conventional linear subspace learning methods, GND-ICA is capable of obtaining more distinctive representation for color image classification.

      CiNii Article

      researchmap

    • Sparse Dictionary Representation and Propagation for MRI Volume Super-Resolution (画像工学)

      Han Xian-Hua, Chen Yen-Wei

      電子情報通信学会技術研究報告. IE, 画像工学112 ( 36 ) 13 - 18   2012年5月10日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      This study addresses the problem of generating a high-resolution (HR) MRI volume from a single low-resolution (LR) MRI input volume. Recent researches have proved that sparse coding can be successfully applied for single-frame super-resolution for natural images, which is based on good reconstruction of any local image patch with a sparse linear combination of atoms taken from an appropriate over-complete dictionary. This study adapts the basic idea of sparse code-based super-resolution (SCSR) for MRI volume data, and then improves the dictionary learning strategy in the conventional SCSR for achieving the precise sparse representation of HR volume patches. In the proposed MRI super-resolution strategy, we only learn the dictionary of the HR MRI volume patches with sparse coding algorithm, and then propagate the HR dictionary to the LR dictionary by mathematical analysis for calculating the sparse representation (coefficients) of any LR local input volume patch. The unknown corresponding HR volume patch can be reconstructed with the sparse coefficients from the LR volume patch and the corresponding HR dictionary. We validate that the proposed SCSR strategy through dictionary propagation can recover much clearer and more accurate HR MRI volume than the conventional interpolated methods.

      CiNii Article

      researchmap

    • ベクトル量子化と多様体学習を用いた超解像技術

      谷口 和輝, 韓 先花, 岩本 祐太郎, 笹谷 聡, 陳 延偉

      電子情報通信学会技術研究報告. MI, 医用画像112 ( 38 ) 19 - 24   2012年5月10日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      近年,画質改善の方法として超解像技術が注目されている.その中でも単一の低解像度画像から高解像度画像を生成するSingle-Frame SRは,入力する低解像度画像とは別に,大規模なデータベースが必要である.しかし,従来まではデータベース内の学習データが冗長となる問題点や,パッチを探索する計算時間が膨大となる問題点があった.そのため,本研究ではk-means法を用いたベクトル量子化によりデータベースの冗長性を低減し,計算時間の高速化を実現した.

      CiNii Article

      researchmap

    • SPHARMベース統計形状モデルと点分布ベース統計形状モデルとの性能比較

      健山 智子, 桶川 萠, 小原 伸哉, 韓 先花, 古川 彰, 金崎 周造, 陳 延偉

      電子情報通信学会技術研究報告. MI, 医用画像112 ( 38 ) 39 - 44   2012年5月10日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      高精細医用画像から構築された人体臓器統計形状モデルは臓器の形状や大きさなどバリエーションを記述したモデルであり,計算解剖学をはじめ診断や治療支援への応用が期待できる.高精度でかつ少数の学習症例数に対しても汎化能力の高い統計形状モデル構築には症例間において解剖学的に同じ特徴の高精度な対応付け,効率的な3次元臓器形状表現確立が求められる.従来は点分布に基づく3次元形状表現に基づいた臓器形状表現による統計形状モデルでは,直交座標系を用いて臓器特徴ベクトル表現,解析を行なっていたが,学習症例に対し,特徴ベクトル次元が膨大になってしまい,汎化能力が低くなるという問題点があった.この解決法として,球座標上で高効率的に臓器形状表現を行うことができる球面調和関数を用いた臓器形状ベクトル表現の確立が注目されている.本研究では,信頼性の高い統計形状モデル構築を目指し,従来のPDM法による統計形状モデルとSPHARMベースの統計形状モデルの比較評価を行う.比較評価は,臓器形状の再現性を示す汎化能力,臓器形状表現の簡潔性,臓器形状特異値排他能力の3つについて検証を行った。

      CiNii Article

      researchmap

    • D-12-22 Visual Impression Estimation of Clothing Fabric Images Using Machine Learning Methods

      Chen Dingye, Han Xianhua, Huang Yang, Huang Xinyin, Chen Yen-wei

      電子情報通信学会総合大会講演論文集2012 ( 2 ) 116 - 116   2012年3月6日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • 球面調和関数を用いた人体脾臓の3次元形状モデリングと統計形状モデルの構築

      健山智子, 小原伸哉, 田中泰史, 桶川萌, 古川顕, 金崎周造, 若宮誠, 韓先花, 陳延偉

      画像の認識・理解シンポジウム(MIRU2011)論文集2011 ( 2011 ) 255 - 262   2011年7月20日

      詳細を見る

      記述言語:日本語  

      CiNii Article

      researchmap

    • Independent components selection of color SIFT descriptors for image classification (パターン認識・メディア理解)

      Ai Dan-ni, Han Xianhua, 段 桂芳, RUAN Xiang, CHEN Yen-Wei

      電子情報通信学会技術研究報告. PRMU, パターン認識・メディア理解111 ( 48 ) 25 - 30   2011年5月12日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      This paper addresses the problem of ordering the color SIFT descriptors in the independent component analysis for image classification. Component ordering is of great importance for image classification, since it is the foundation of the feature selection. To select the distinctive and compact independent components of the color SIFT descriptors, we propose two ordering approaches based on local variation, named as the localization-based ICs ordering and the sparseness-based ICs ordering. We evaluate the performance of proposed methods, the conventional ICs selection method (global variation based components selection) and original color SIFT descriptors on object and scene databases, and obtain the following two main results. First, the proposed methods are able to obtain the acceptable classification results in comparison with original color SIFT descriptors. Second, the highest classification rate can be obtained by using the global selection method in the scene database, while the local ordering methods give the best performance for the object database.

      CiNii Article

      researchmap

    • Joint kernel equal integration of visual features and textual terms for biomedical imaging modality classification (画像工学)

      韓 先花, 陳 延偉

      電子情報通信学会技術研究報告. IE, 画像工学111 ( 47 ) 51 - 56   2011年5月12日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      In this paper, we describe an approach for the automatic modality classification in medical image retrieval task of the 2010 CLEF cross-language image retrieval campaign (ImageCLEF). This work is focused on the process of feature extraction from medical images and fusion the different extracted visual feature and textual feature for modality classification. To extract visual features from the images, we used histogram descriptor of edge, gray or color intensity and block-based variation as global features and SIFT histogram as local feature, and the binary histogram of some predefined vocabulary words for image captions is used for textual feature. Then we combine the different features using normalized kernel functions for SVM classification. The proposed algorithm is evaluated by the provided modality dataset by ImageCLEF2010.

      CiNii Article

      researchmap

    • Independent components selection of color SIFT descriptors for image classification (画像工学)

      Ai Dan-ni, Han Xianhua, 段 桂芳, RUAN Xiang, CHEN Yen-Wei

      電子情報通信学会技術研究報告. IE, 画像工学111 ( 47 ) 25 - 30   2011年5月12日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      This paper addresses the problem of ordering the color SIFT descriptors in the independent component analysis for image classification. Component ordering is of great importance for image classification, since it is the foundation of the feature selection. To select the distinctive and compact independent components of the color SIFT descriptors, we propose two ordering approaches based on local variation, named as the localization-based ICs ordering and the sparseness-based ICs ordering. We evaluate the performance of proposed methods, the conventional ICs selection method (global variation based components selection) and original color SIFT descriptors on object and scene databases, and obtain the following two main results. First, the proposed methods are able to obtain the acceptable classification results in comparison with original color SIFT descriptors. Second, the highest classification rate can be obtained by using the global selection method in the scene database, while the local ordering methods give the best performance for the object database.

      CiNii Article

      researchmap

    • Hallucination 法に基づいた医用画像の超解像処理

      岩本 祐太郎, 韓 先花, 大橋 基範, 笹谷 聡, 谷口 和輝, 陳 延偉

      電子情報通信学会技術研究報告. IE, 画像工学111 ( 47 ) 57 - 62   2011年5月12日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      医用画像は,撮影条件などの影響により医療診断を行う上で必ずしも十分な解像度の画像を得られるとは限らない.特に断層間(Z方向)の解像度は断層面(XY方向)の解像度に比べ一般的に低い傾向にある.そのため手術シミュレーションなどの三次元可視化システムにおいて重要となる等方性ボクセル(縦横サイズの等しいボクセル)を構成することが困難であるといった問題点も挙げられる.そのため,本論文では学習型超解像技術の一種であるHallucination法を基に,断層間(Z方向)の高解像度化を図る手法を提案する.そして従来の補間手法との比較を行い,提案手法の有用性を検証する.

      CiNii Article

      researchmap

    • SPIN特徴量を用いた食事画像の自動認識と食事ログシステム

      和澄 南, 鶴貝 陽平, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告. IE, 画像工学111 ( 47 ) 93 - 98   2011年5月12日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      近年,食生活の乱れや健康的な生活意識の向上に伴い,食生活の管理が必要となっている.そこで携帯電話のカメラ機能と用いて撮影された食事画像から自動で食事内容の認識を行なうことで,よりユーザに身近なシステムの構築を目指している.しかしここで問題として,食事画像内の食品の向きが一定でないということが挙げられる.その解決策として,本稿ではSPINと呼ばれる円形分割を用いることで回転に不変な特徴量を抽出し,食品の認識を行なっている.また,認識結果から食事内容や摂取エネルギー・栄養素を記録する食事ログを作成することで,自らの食事状況を確認でき,食事管理に役立てることを可能としている.

      CiNii Article

      researchmap

    • 高周波成分を考慮したFace Hallucination法 (コンピュータビジョンとイメージメディア(CVIM) Vol.2011-CVIM-175)

      笹谷 聡, 韓 先花, 大橋 基範

      情報処理学会研究報告2010 ( 5 ) 6p   2011年2月

      詳細を見る

      記述言語:日本語   出版者・発行元:情報処理学会  

      CiNii Article

      researchmap

    • 高周波成分を考慮した Face Hallucination 法

      笹谷 聡, 韓 先花, 大橋 基範, 岩本 祐太郎, 陳 延偉

      電子情報通信学会技術研究報告. MVE, マルチメディア・仮想環境基礎110 ( 382 ) 323 - 328   2011年1月13日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      学習べース超解像処理の1つであるFace Hallucination法は,事例学習に基づき単一低解像度顔画像から高解像度顔画像を生成する手法である.この手法により生成した画像は高周波成分が欠落し本来の顔特徴が失われている可能性があり,何らかの方法で高周波成分を復元しなければならない.本発表では,高周波成分を考慮したFace Hallucination法を提案する.提案法は,以下の3つのステップからなる.(1).従来のFace Hallucination法により高解像度画像を生成する: (2).生成した高解像度画像をダウンサンプリングし,入力低解像度画像との残差画像を生成し,その残差画像から従来のFace Hallucination法で高解像度残差画像を生成する;(3).ステップ1で生成した高解像度画像にステップ 2で生成した高解像度残差画像を補填することでより鮮明な高解像度顔画像を生成する.

      CiNii Article

      researchmap

    • 回転不変位相限定相関法の超解像技術への応用(インタラクティブシステム・画像入力デバイス・方式,及び一般)

      谷口 和輝, 韓 先花, 岩本 祐太郎, 笹谷 聡, 大橋 基範, 陳 延偉

      映像情報メディア学会技術報告35 ( 0 ) 33 - 36   2011年

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人 映像情報メディア学会  

      画像の解像度をソフトウェア面から向上させる超解像技術の中で,複数枚の低解像度画像を用いて一枚の高解像度画像を生成するマルチフレーム手法があるが,この手法では大きな回転を含む画像群を使用することは困難であった.しかし,リモート・センシングなどの分野において回転を含む画像は多く存在している.この問題点を改善するべく,我々は位置合わせ手法の一つである回転不変位相限定相関法によって得られた回転パラメータに対して勾配法を用いることで高精度な推定を可能にし,高解像度化を図る手法を提案する.本論文では,提案法に反復逆投影法を加えることで,より鮮明な画像を生成し,補間拡大画像との比較から提案手法の有用性を検証する.

      DOI: 10.11485/itetr.35.23.0_33

      CiNii Article

      researchmap

    • 多重解像度セマンティックモデルを用いた画像のシーン認識

      田中 良幸, 岡本 充史, 韓 先花, 阮 翔, 陳 延偉

      電子情報通信学会技術研究報告. MI, 医用画像110 ( 28 ) 169 - 174   2010年5月6日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      計算機による画像認識を実現することにより画像検索等で応用が期待され,様々な手法が提案されている.画像認識には物体自体を認識する他に,風景を認識するシーン認識という分野が存在する.近年シーン認識の手法に,分割した局所画像から画像全体を認識するセマンティックモデルが提案された.しかしこの手法には静的なグリッドを用いるため幅広いシーン画像には対応しきれていないという問題がある.本研究ではそれを改良し,複数サイズの局所画像を用いる多重解像度セマンティックモデルという手法について述べる.この手法は分類結果が信頼できるかどうかを計算し,その結果をもとに最適なグリッド数を選択,分類する手法である.その手法と従来の手法との比較実験を行い,提案手法が優れていることを示す.

      CiNii Article

      researchmap

    • 複数特徴量の融合および階層的分類法によるシーン認識

      菊谷 八重, 岡本 充史, 韓 先花, 阮 翔, 陳 延偉

      電子情報通信学会技術研究報告. MI, 医用画像110 ( 28 ) 175 - 179   2010年5月6日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      近年デジタルカメラの機能の一部として自動シーン認識が注目されている.自動シーン認識はコンピュータビジョンにおいて高度な処理であり,未解決の困難な問題である.従来の手法では一般的にカラーヒストグラムやテクスチャ,エッジなどの大域的な特徴量を使用して画像を表現し,ベイズ法やニューラルネットワーク,SVMなどの分類器を用いてどのシーンであるかを認識している.しかし,実用アプリケーションとして認識率は十分とは言えない.そこで本研究の提案法では大域的な特徴量であるカラーヒストグラムと局所的な特徴量であるBag of Featuresを重み付き融合し,階層的分類法により分類器をシーンの特性に最適化する.提案法による実験では認識率の改善を示すことができた.

      CiNii Article

      researchmap

    • 主成分分析と勾配法を併用した位置合わせと超解像技術

      笹谷 聡, 韓 先花, 陳 延偉

      電子情報通信学会技術研究報告. MI, 医用画像110 ( 28 ) 193 - 198   2010年5月6日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      解像度を上げる技術である超解像技術には,複数枚の低解像度画像から1枚の高解像度画像を生成するマルチフレーム手法がある.この手法で使用される画像間の位置合わせ技術には様々な種類があるが,各々に対して問題点がありこれらを改善しより正確な位置合わせを行うことによって,生成する画像のさらなる画質向上が可能となる.本発表では既存手法を併用した新しい位置合わせ技術を紹介し,その実験結果について述べる.

      CiNii Article

      researchmap

    • An adaptive color SIFT descriptor based on independent component analysis for image classification (パターン認識・メディア理解)

      Ai Danni, Han Xianhua, Ruan Xiang, CHEN Yen-Wei

      電子情報通信学会技術研究報告. PRMU, パターン認識・メディア理解109 ( 306 ) 1 - 6   2009年11月19日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      In this paper, we present a novel color independent components based SIFT descriptor (termed CIC-SIFT) for object/scene classification. We first learn an efficient color transformation matrix based on independent component analysis (ICA), which is adaptive to each category in a database. The ICA-based color transformation can enhance contrast between the objects and the background in an image. Then we compute CIC-SIFT descriptors over all three transformed color independent components. Since the ICA-based color transformation can boost the objects and suppress the background, the proposed CIC-SIFT can extract more effective and discriminative local features for object/scene classification. The comparison is performed among seven SIFT descriptors and the experimental classification results show that our proposed CIC-SIFT is superior to other conventional SIFT descriptors.

      CiNii Article

      researchmap

    • Multi-class object recognition by fusion of image descriptors: classification evaluation of PASCAL VOC Challenge database (パターン認識・メディア理解)

      Han Xian-Hua, Chen Yen-Wei, Ruan Xiang

      電子情報通信学会技術研究報告. PRMU, パターン認識・メディア理解109 ( 306 ) 103 - 108   2009年11月19日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Category recognition is important to access visual information on the level of objects. Over the past several years, substantial performance gains on challenging benchmark datasets have been reported in the literature. This progress can be attributed to two developments: the design of highly discriminative and robust image features and the combination of multiple complementary features based on different aspects such as shape, appearance. In this paper we extract different appearance descriptors for several properties of local keypionts obtained by grid sampling, Harris and Difference of Gaussian (DoG), and a shape descriptor-Pyramid Histogram of Orientation Gradient (PHOG). The extracted descriptors are evaluated by PASCAL VOC database with nonlinear SVM classifiers, and they also are combined for evaluation. The experimental results show that the accuracy rate with fusion of different image descriptors can be improved.

      CiNii Article

      researchmap

    • A supervised LPP and neural network based scene classification with color histogram and camera metadata (パターン認識・メディア理解)

      Han Xian-Hua, Chen Yen-Wei, Fujita Hideto, KOYAMA Kan-Ichi, TAKAYANAKI Wataru

      電子情報通信学会技術研究報告. PRMU, パターン認識・メディア理解108 ( 484 ) 257 - 262   2009年3月6日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Scene classification (e.g., landscape, sunset, night-landscape, etc.) is still a challenging problem in computer vision. Scene classification based only on low-level vision cues has had limited success on unconstrained image sets. In other hand, camera metadata related to capture conditions provides cues independent of the captured scene content that can be used to improve classification performance. Analysis of camera metadata statistics for images of each class revealed that some metadata fields are most discriminative for some classes. So, in this paper, we proposed to use the combined feature of scene color histogram and camera metadata, and then using supervised Locality preserving projection (LPP) for feature space transformation and dimension reduction, and finally, adapt Probabilistic neural network for scene classification. Experimental results show that the classification accuracy rate can be improved compared with using PCA (Principal Component Analysis) subspace learning method, and are also better than that with only the low-level vision feature (color histogram).

      CiNii Article

      researchmap

    • A supervised LPP and neural network based scene classification with color histogram and camera metadata (コンピュータビジョンとイメージメディア)

      Xian-HuaHan, Yen-WeiChen, Hideto Fujita, Kan-IchiKoyama, Wataru Takayanaki

      研究報告コンピュータビジョンとイメージメディア(CVIM)2009 ( 29 ) 257 - 262   2009年3月6日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人情報処理学会  

      Scene classification (e.g., landscape, sunset, night-landscape, etc.) is still a challenging problem in computer vision. Scene classification based only on low-level vision cues has had limited success on unconstrained image sets. In other hand, camera metadata related to capture conditions provides cues independent of the captured scene content that can be used to improve classification performance. Analysis of camera metadata statistics for images of each class revealed that some metadata fields are most discriminative for some classes. So, in this paper, we proposed to use the combined feature of scene color histogram and camera metadata, and then using supervised Locality preserving projection (LPP) for feature space transformation and dimension reduction, and finally, adapt Probabilistic neural network for scene classification. Experimental results show that the classification accuracy rate can be improved compared with using PCA (Principal Component Analysis) subspace learning method, and are also better than that with only the low-level vision feature (color histogram).Scene classification (e.g., landscape, sunset, night-landscape, etc.) is still a challenging problem in computer vision. Scene classification based only on low-level vision cues has had limited success on unconstrained image sets. In other hand, camera metadata related to capture conditions provides cues independent of the captured scene content that can be used to improve classification performance. Analysis of camera metadata statistics for images of each class revealed that some metadata fields are most discriminative for some classes. So, in this paper, we proposed to use the combined feature of scene color histogram and camera metadata, and then using supervised Locality preserving projection (LPP) for feature space transformation and dimension reduction, and finally, adapt Probabilistic neural network for scene classification. Experimental results show that the classification accuracy rate can be improved compared with using PCA (Principal Component Analysis) subspace learning method, and are also better than that with only the low-level vision feature (color histogram).

      CiNii Article

      researchmap

      その他リンク: http://id.nii.ac.jp/1001/00061527/

    • D-12-36 A Spatial Weighted Edge Autocorrelogram for Image Annotation

      Ai Danni, Duan Guifang, Han Xianhua, Ruan Xiang, Chen Yen-Wei

      電子情報通信学会総合大会講演論文集2009 ( 2 ) 145 - 145   2009年3月4日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • D-12-18 PCA-CSIFTによる自動Annotation付与(D-12.パターン認識・メディア理解A(パターンメディアの認識・理解・生成),一般セッション)

      岡本 充史, 韓 先花, 阮 翔, 陳 延偉

      電子情報通信学会総合大会講演論文集2009 ( 2 ) 127 - 127   2009年3月4日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • D-12-17 高倍率画像を生成するための階層的Super Resolution(D-12.パターン認識・メディア理解A(パターンメディアの認識・理解・生成),一般セッション)

      大橋 基範, 韓 先花, 陳 延偉

      電子情報通信学会総合大会講演論文集2009 ( 2 ) 126 - 126   2009年3月4日

      詳細を見る

      記述言語:日本語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • 改善された独立成分分析の手法に基づいたロバスト顔認識

      韓 先花, 陳 延偉, 山田 晶彦, 藤田 日出人

      情報処理学会研究報告. CVIM, [コンピュータビジョンとイメージメディア]164 ( 82 ) 79 - 84   2008年8月29日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人情報処理学会  

      顔認識を実現するためのプログラムでは、統計的パターン認識手法が重要な役割を演じる。その中には、主成分分析(PCA)、独立成分分析(ICA)と線形判別分析(LDA)等部分空間学習アルゴリズムを用いた顔認識の研究、開発が活発に行われている。LDAは教師付き顔認証のために良い結果を得ることができます、しかし、認識するsubjectsのトレーニング顔はなしあるいは1つしか手入れできないのシステムの応用が制限されます。そして、PCAによる顔画像表現に比べICAによる顔画像表現の有効性を議論した。しかし、顔認証のような次元数が多すぎなパターン認識作業へのICAの適用性は、しばしば若干の問題で苦しみます。その問題の一つは次元数が膨大となり、計算量が大きくとなる、もう一つはコンピュータメモリの問題です。本稿では、ICAのこの2つの問題を扱うために、独立成分分析の改善されたarchitectureを提案する。提案されたICA手法は、顔認識が高速に行われていることができるだけではなく、実験の結果により顔認識の有効性が確認することができた。

      CiNii Article

      researchmap

    • 改善された独立成分分析の手法に基づいたロバスト顔認識

      韓 先花, 陳 延偉, 山田 晶彦, 藤田 日出人

      電子情報通信学会技術研究報告. HIP, ヒューマン情報処理108 ( 199 ) 79 - 84   2008年8月29日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      顔認識を実現するためのプログラムでは、統計的パターン認識手法が重要な役割を演じる。その中には、主成分分析(PCA)、独立成分分析(ICA)と線形判別分析(LDA)等部分空間学習アルゴリズムを用いた顔認識の研究、開発が活発に行われている。LDAは教師付き顔認証のために良い結果を得ることができます、しかし、認識するsubjectsのトレーニング顔はなしあるいは1つしか手入れできないのシステムの応用が制限されます。そして、PCAによる顔画像表現に比べICAによる顔画像表現の有効性を議論した。しかし、顔認証のような次元数が多すぎなパターン認識作業へのICAの適用性は、しばしば若干の問題で苦しみます。その問題の一つは次元数が膨大となり、計算量が大きくとなる、もう一つはコンピュータメモリの問題です。本稿では、ICAのこの2つの問題を扱うために、独立成分分析の改善されたarchitectureを提案する。提案されたICA手法は、顔認識が高速に行われていることができるだけではなく、実験の結果により顔認識の有効性が確認することができた。

      CiNii Article

      researchmap

    • D-16-2 ICA Based Ring Artifacts Reduction in Cone-Beam CT Images

      DUAN Guifang, CHEN Yen-wei, HAN Xianhua, FUJITA Akinori, HIROOKA Ken, UENO Yoshihiro

      電子情報通信学会総合大会講演論文集2008 ( 2 ) 269 - 269   2008年3月5日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      CiNii Article

      researchmap

    • Reduction of ring artifacts in cone-beam CT images (医用画像)

      Duan Guifang, Chen Yen-wei, Han Xianhua, FUJITA Akinori, HIROOKA Ken, UENO Yoshihiro

      電子情報通信学会技術研究報告. MI, 医用画像107 ( 461 ) 227 - 233   2008年1月25日

      詳細を見る

      記述言語:英語   出版者・発行元:一般社団法人電子情報通信学会  

      Cone-beam CT (CBCT) scanners are based on volumetric tomography, using a 2D extended digital array providing an area detector [1]. Compared to traditional CT, CBCT has many advantages, such as less X-ray beam limitation, high image accuracy, Rapid scan time, etc. However, In CBCT images there are always some ring artifacts that appear as rings centred on the rotation axis. Due to the data of the constructed images are corrupted by these ring artifacts, qualitative and quantitative analysis of CBCT images will be compromised. Post processing and application such as image segmentation and registration also turn more complex as the presence of such artifacts. In this paper, a method with a novel region of interest (ROI) building technique is presented. It deals with the reconstructed CBCT image and can effectively reduce such ring artifacts.

      CiNii Article

      researchmap

    ▼全件表示

    書籍等出版物

    • Residual Sparse Autoencoders for Unsupervised Feature Learning and Its Application to HEp-2 Cell Staining Pattern Recognition

      Xian-Hua Han, Yen-Wei Chen( 担当: 分担執筆 ,  範囲: Chapter 11)

      Deep Learning in Healthcare, Chapter 11, Yen-Wei Chen and Jain, Lakhmi C. (Eds.), [Online First], Springer, 2020. DOI: 10.1007/978-3-030-32606-7_11  2020年 

      詳細を見る

    • Hyperspectral Image Super-Resolution Using Optimization and DCNN-Based Methods

      Xian-Hua Han( 担当: 分担執筆 ,  範囲: One chapter)

      Processing and Analysis of Hyperspectral Data, Jie Chen (Eds.), [Online First], IntechOpen, DOI: 10.5772/intechopen.89243.  2020年 

      詳細を見る

    • High-Order Statistics of Micro-Texton for HEp-2 Staining Pattern Classification in 「Artificial Intelligence in Decision Support Systems for Diagnosis in Medical Imaging」

      Xian-Hua Han, Yen-Wei Chen( 担当: 分担執筆 ,  範囲: Chapter 4)

      Springer 出版  2018年1月 

      詳細を見る

    • Subspace Methods for Pattern Recognition in Intelligent Environment・Sparse Representation for Image Super-Resolution

      韓 先花, 陳 延偉( 範囲: 123-150)

      Springer  2014年6月  ( ISBN:9783642548505

      詳細を見る

    • Adaptive Noise Reduction and Edge Enhancement in Medical Images by using ICA?Computational Intelligence in Biomedical Imaging

      Xian-Hua Han, Yen-Wei chen

      Springer  2013年6月  ( ISBN:9781461472445

      詳細を見る

    • Multilinear Supervised Neighborhood Preserving Embedding Analysis of Local Descriptor Tensor?Principal Component Analysis

      Xian-Hua Han, Yen-wei Chen( 範囲: 91-106)

      INTECH  2012年3月  ( ISBN:9789535101956

      詳細を見る

    ▼全件表示

    講演・口頭発表等

    • 正規化スパース性制約を用いたぼけカーネル推定および動画像復元への応用

      野島優輔, 韓先花, 陳延偉

      信学技報, vol. 115, no. 24, PRMU2015-1 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • Robust Blood Vessel of Liver Extraction Using Deep Convolution Network

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yutaro Iwamoto, Yen-Wei Chen

      信学技報, vol. 116, no. 225, MI2016-48 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • 深層Residual Attention ネットワークを用いたスナップショット撮影からハイパースペクトル画像の復元

      寄元康平, 韓先花

      第19回情報科学技術フォーラム(FIT2020), H-006 

      詳細を見る

      開催年月日: 2020年9月1日 - 2020年9月3日

      会議種別:口頭発表(一般)  

      researchmap

    • Deep Attention ConvLSTM-UNetを用いた画像の降雨ノイズ除去

      山道航平, 韓先花

      第19回情報科学技術フォーラム(FIT2020), H-028 

      詳細を見る

      開催年月日: 2020年9月1日 - 2020年9月3日

      会議種別:口頭発表(一般)  

      researchmap

    • 注目領域の自動学習ネットワークによる画像キャプションの自動生成

      寄元康平, 韓先花

      第23回 画像の認識・理解シンポジウム (MIRU2020) 

      詳細を見る

      開催年月日: 2020年8月3日 - 2020年8月5日

      会議種別:ポスター発表  

      researchmap

    • 知覚損失学習を用いた単一画像の雨筋除去

      山道航平, 韓先花

      第23回 画像の認識・理解シンポジウム (MIRU2020) 

      詳細を見る

      開催年月日: 2020年8月3日 - 2020年8月5日

      会議種別:ポスター発表  

      researchmap

    • Multispectral and Hyperspectral Image Fusion with Deep Learned Prior

      Zhe Liu, Xian-Hua Han

      第23回 画像の認識・理解シンポジウム (MIRU2020) 

      詳細を見る

      開催年月日: 2020年8月3日 - 2020年8月5日

      会議種別:ポスター発表  

      researchmap

    • Deeply Supervised ConvLSTM-Unetによる医療用画像の分割

      方文昊, 韓先花

      第23回 画像の認識・理解シンポジウム (MIRU2020) 

      詳細を見る

      開催年月日: 2020年8月3日 - 2020年8月5日

      会議種別:ポスター発表  

      researchmap

    • 段階的逆投影ネットワークによるハイパースペクトル画像の高解像度化

      吉武大輝, 韓先花

      第23回 画像の認識・理解シンポジウム (MIRU2020) 

      詳細を見る

      開催年月日: 2020年8月3日 - 2020年8月5日

      会議種別:ポスター発表  

      researchmap

    • 画像のCNN特徴とLSTMのデンス結合構造を用いた画像キャプションの生成

      寄元康平, 韓先花

      2020年電子情報通信学会総合大会, D-12-42 

      詳細を見る

      開催年月日: 2020年3月17日 - 2020年3月20日

      会議種別:口頭発表(一般)  

      researchmap

    • 深層知覚損失ネットワークを用いた画像の降雨ノイズ除去

      山道航平, 韓先花

      2020年電子情報通信学会総合大会, D-12-59 

      詳細を見る

      開催年月日: 2020年3月17日 - 2020年3月20日

      会議種別:口頭発表(一般)  

      researchmap

    • 改良したSSDを用いた航空写真からsmall scare物体の検出と分類

      中井克啓, 韓 先花

      2020年電子情報通信学会総合大会, D-12-61 

      詳細を見る

      開催年月日: 2020年3月17日 - 2020年3月20日

      会議種別:口頭発表(一般)  

      researchmap

    • Mask R-CNNを用いた細胞核の検出とセグメンテーション

      藤田 世哉, 韓 先花

      2020年電子情報通信学会総合大会, A-15-5 

      詳細を見る

      開催年月日: 2020年3月17日 - 2020年3月20日

      会議種別:口頭発表(一般)  

      researchmap

    • Deep supervision U-netとconvLSTMの統合による医療画像の分割

      方 文昊, 韓 先花

      2020年電子情報通信学会総合大会, A-15-6 

      詳細を見る

      開催年月日: 2020年3月17日 - 2020年3月20日

      会議種別:口頭発表(一般)  

      researchmap

    • Hyperspectral Residual Component Reconstruction with Deep ResNet

      Xian-Hua Han, Hiroki Yoshitake, Yen-Wei Chen

      第22回 画像の認識・理解シンポジウム (MIRU2019)  2019年8月1日 

      詳細を見る

      記述言語:英語   会議種別:ポスター発表  

      researchmap

    • 敵対的生成ネットワークを用いたハイパースペクトル画像の高解像度化

      吉武 大輝, 韓 先花

      第22回 画像の認識・理解シンポジウム (MIRU2019)  2019年7月31日 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • Region CNNを用いた細胞核の検出

      藤田 世哉, 韓 先花

      第22回 画像の認識・理解シンポジウム (MIRU2019)  2019年7月31日 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • Computer-Aided Diagnosis of Liver Cancers Using Deep Learning with Fine-tuning

      Weibin WANG, Dong LIANG, Lanfen LIN, Hongjie HU, Qiaowei ZHANG, Qingqing CHEN, Yutaro IWAMOTO, Xianhua HAN, Yen-Wei CHEN

      電子情報通信学会PRMU研究会, 福岡工業大学  2018年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • 深層学習技術を用いたCT画像からの肝臓腫瘍候補の検出

      轟佳大, 韓先花, 岩本祐太郎, Lanfen Lin, Hongjie Hu, 陳延偉

      第8回横幹連合コンファレンス, P14-S, 京都  2017年12月2日 

      詳細を見る

    • Deep Convolutional Neural Networkを用いた食事画像認識

      佐藤亮輔, 韓先花, 岩本祐太郎, 陳延偉

      平成29年電気関係学会関西連合大会  2017年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • Automatic Vessel Segmentation Using A combined Deep Network

      Titinunt KITRUNGROTSAKUL, Xian-Hua Han, Yutaro Iwamoto, Yen-Wei Chen

      第36回日本医用画像工学会大会(JAMIT2017).  2017年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • Learning an overcomplete codebook of tensor local structures for multi-phase medical image retrieval

      Jian Wang, Xian-Hua Han, Yingying Xu, Lanfen Lin, Hongjie Hu, Chongwu Jin, Yen-Wei Chen

      第36回日本医用画像工学会大会(JAMIT2017).  2017年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • スパース自己符号化器を用いてHEp-2 細胞画像認識システム

      韓 先花, 陳 延偉

      第36回日本医用画像工学会大会(JAMIT2017).  2017年 

      詳細を見る

      会議種別:口頭発表(一般)  

      researchmap

    • Robust Hyperspectral Image Super-Resolution via Constrained Non-negative Sparse Representation

      Xian-Hua Han, Boxin Shi, Yinqiang Zheng, Toru Kouyama, Atsunori Kanemura, Ryosuke Nakamura

      画像の認識・理解シンポジウム (MIRU2017).  2017年 

      詳細を見る

      会議種別:ポスター発表  

      researchmap

    • 深層学習を用いた肝臓腫瘍候補の検出

      轟佳大,TitinuntKitrungrotsakul, 岩本祐太郎, 韓先花, 陳延偉

      平成28年度電気関係学会関西連合大会  2016年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • A preliminary study on tensor codebook model for multiphase medical image retrieval

      Jian Wang, Xian-Hua Han, Yingying Xu, Lanfen Lin, Hongjie Hu, Chongwu Jin, Yen-Wei Chen

      信学技報, vol. 116, no. 225, MI2016-58  2016年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • HEp-2 cell classification using K-support spatial pooling in Deep CNNs

      Xian-Hua Han, Yen-Wei Chen

      信学技報, vol. 116, no. 225, MI2016-47  2016年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • BoVWとBoTWを用いた肝腫瘍CT画像の検索

      Yingying XU, Lanfen LIN, Hongjie HU, Dan WANG, Yitao LIU, Jian WANG, Xianhua HAN, Yen-Wei CHEN

      35回日本医用画像工学会大会  2016年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • Extraction of Liver Vessel using Graph Cut based Submodular on 3D data

      Titinunt Kitrungrotsakul, Xian-Hua Han, Yen-Wei Chen

      第35回日本医用画像工学会大会  2016年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • Combined Bags of Color and Texture Featuresを用いた食事画像認識

      笹野翔太, 韓先花, 陳延偉

      第19回画像の認識・理解シンポジウム(MIRU2016),  2016年 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • 三次元脳MR画像の超解像処理とセグメンテーションへの応用

      岩本祐太郎, 韓先花, 椎野顯彦, 陳延偉

      第19回画像の認識・理解シンポジウム(MIRU2016),  2016年 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • Improved Spatial Pyramid Pooling in Deep Convolutional Network for Remote Sensing Image Classification

      Xian-Hua Han, Ryosuke Nakamura, Yen-Wei Chen

      第19回画像の認識・理解シンポジウム  2016年 

      詳細を見る

      記述言語:英語   会議種別:ポスター発表  

      researchmap

    • Super-Resolution for 3D Anisotropic Low-resolution MR Images Using Multilayer Convolutional Neural Networks

      Qiaochu Zhao, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen

      第19回画像の認識・理解シンポジウム(MIRU2016)  2016年 

      詳細を見る

      記述言語:英語   会議種別:ポスター発表  

      researchmap

    • A preliminary study on multilinear sparse coding for medical image retrieval

      Jian Wang, Xian-Hua Han, Yingying Xu, Lanfen Lin, Hongjie Hu, Chongwu Jin, Yen-Wei Chen

      第19回画像の認識・理解シンポジウム(MIRU2016)  2016年 

      詳細を見る

      記述言語:英語   会議種別:ポスター発表  

      researchmap

    • 2段階Hallucination法による医用画像の高解像度化と3次元医用画像への応用

      近藤佑斗, 韓先花, 陳延偉

      2015年電子情報通信総合大会,D-11-34, 滋賀県草津  2015年 

      詳細を見る

    • Sparse and Low Rank DecompositionとICPを用いた頑健な点群位置合わせ法

      趙きょうそ, 韓先花, 陳延偉

      信学技報, vol. 115, no. 24, PRMU2015-5  2015年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • Liver Segmentation Using Superpixel-based Graph cuts and Shape Constraints

      TitinuntKitrungrotsakul, Xian-Hua Han, Yen-Wei Chen

      第18回画像の認識・理解シンポジウム  2015年 

      詳細を見る

      記述言語:英語   会議種別:ポスター発表  

      researchmap

    • ベイズモデルを用いたCT画像からの腫瘍候補の検出

      今野悠, 韓先花, Xiong Wei, 陳延偉

      第18回画像の認識・理解シンポジウム  2015年 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • Sparse and Low Rank DecompositionとHuber ICPを用いた頑健な点群位置合わせ法

      趙きょうそ, 韓先花, 陳延偉

      第18回画像の認識・理解シンポジウム  2015年 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • 3次元顔面形態の局所的統計解析

      中津美冴, 韓先花, 木村亮介, 陳延偉

      第18回画像の認識・理解シンポジウム  2015年 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • データ駆動モデルを用いた食事画像識別

      笹野翔太, 中島基輝, 韓先花, 陳延偉

      第18回画像の認識・理解シンポジウム  2015年 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    • Deep Convolutional Neural Network による食事画像認識

      韓先花, 陳延偉

      信学技報, vol. 115, no. 224, PRMU2015-77  2015年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • Generalized Super-Vectorを用いた一般画像分類

      中島基輝, 韓 先花, 陳 延偉

      信学技報, vol. 114, no. 41, PRMU2014-21  2014年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • ノンパラメトリック確率モデルを用いたCT画像からの腫瘍候補の検出

      今野 悠, 韓 先花, 陳 延偉

      信学技報, vol. 114, no. 41, PRMU2014-14  2014年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • 統計学習法を用いた3次元顔面形態解析

      中津美冴, 韓 先花, 木村亮介, 陳 延偉

      信学技報, vol. 114, no. 41, PRMU2014-13  2014年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • 2段階Hallucination法および3次元医用画像の高解像度化への応

      近藤佑斗, 野島優補, 韓 先花, 陳 延偉

      信学技報, vol. 114, no. 41, PRMU2014-12  2014年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • Multi-touch Based Medical Interactive Visualization System

      Jian Wang, Hua-Wei Tu, Xian-Hua Han, Tomoko Tateyama, Yen-Wei Chen

      信学技報, vol. 114, no. 41, PRMU2014-7  2014年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • スパースコーディングとハイブリッドプーリングを用いた食事画像認識

      楠元理子, 韓 先花, 陳 延偉

      信学技報, vol. 114, no. 41, PRMU2014-1  2014年 

      詳細を見る

      記述言語:日本語   会議種別:口頭発表(一般)  

      researchmap

    • Organ Bounding Box Annotation based on Adaptive Selection of Bone References

      Chunhua Dong・Amir, H. Foruzan, Xian-hua Han, Tomoko Tateyama, Yen-wei Chen

      信学技報, vol. 114, no. 103, MI2014-27  2014年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • Improved Interactive Medical Image Segmentation using Graph Cut and Superpixels

      Titinunt Kitrungrotsakul・Chunhua Dong, Xian-Hua Han, Yen-Wei Chen

      信学技報, vol. 114, no. 103, MI2014-25  2014年 

      詳細を見る

      記述言語:英語   会議種別:口頭発表(一般)  

      researchmap

    • Data-Driven Model of Weber Local Descriptors for Visual Recognition

      Xian-Hua Han, Yen-Wei Chen, Gang Xu

      MIRU2014 第17回画像の認識・理解シンポジウム  2014年 

      詳細を見る

      記述言語:英語   会議種別:ポスター発表  

      researchmap

    • Generalized Super-Vector

      中島 基輝, 韓 先花, 陳 延偉

      MIRU 2014第17回画像の認識・理解シンポジウム  2014年 

      詳細を見る

      記述言語:日本語   会議種別:ポスター発表  

      researchmap

    ▼全件表示

    担当経験のある科目(授業)

    • 2023年4月 - 2023年8月 
      応用機械学習特論 ( 山口大学 )

      詳細を見る

    • 2023年4月 - 2023年8月 
      機械学習 ( 山口大学 )

      詳細を見る

    • 2023年4月 - 2023年8月 
      パターン認識 ( 山口大学 )

      詳細を見る

    • 2022年4月 - 2023年8月 
      データサイエンス技術Ⅱ ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2023年8月 
      機械学習特論 ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2023年8月 
      プログラミング演習Ⅰ ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2023年8月 
      データ科学と社会Ⅱ ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2023年8月 
      データ科学と社会Ⅰ ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2023年8月 
      物理と情報のための基礎数学Ⅲ演習 ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2023年8月 
      基礎セミナー ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2023年3月 
      情報処理特論 ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2021年3月 
      グラフ理論 ( 山口大学 )

      詳細を見る

    • 2017年4月 - 2020年3月 
      自然科学Ⅱ ( 山口大学 )

      詳細を見る

    ▼全件表示

    所属学協会

    共同研究・競争的資金等の研究

    • 知識・データ融合型枠組の開発およびハイパースペクトル画像の超解像度への応用

      日本学術振興会  科学研究費助成事業 基盤研究(C) 

      韓 先花, 陳 延偉, 末竹 規哲

      詳細を見る

      2020年4月 - 2023年3月

      課題番号:20K11867

      配分額:4290000円 ( 直接経費:3300000円 、 間接経費:990000円 )

      本研究では最先端の計測技術を搭載したハイパースペクトル(HS)カメラであっても計測が難しい高解像度なHS画像を生成することを目的に、画像のイメージング過程を定式化する数理モデルと深層学習技術を融合した新たな方法論を確立し、高解像度HS画像を復元する方法を開発する。本年度では、研究実績は主に以下の3点である.
      1)空間構造が異なる二つのモダリティデータを効率的に融合する新たな深層ネットワーク構造を提案し、伝統的な数理モデルに基づく再構成法と統合することで高精度な高解像度HS画像の復元を確認した。
      2)高精細な空間構造と細かいスペクトル分布を同時に復元できる軽量なHyperMixNetフレームワークを提案し、圧縮されたSnapshot画像から高精度なスペクトル情報の復元だけではなく、深層モデルのサイズと計算コストの削減も実現した。
      3)膨大なラベル(高解像度HR画像)付きトレーニングサンプルが必要なし、観測の低解像度HS画像と高解像度RGB画像のみを用いた教師なし深層学習ネットワーク枠組を開発し、高精度なHS画像の生成を実現した。また、計測データの劣化過程を自動的に学習できる汎用性が高いBlind教師なし深層モデルを構築している。

      researchmap

    • テンソルスパース表現による多時相CT画像の時空間特徴抽出と肝腫瘍性病変の診断支援

      日本学術振興会  科学研究費助成事業 基盤研究(B) 

      陳 延偉, 岩本 祐太郎, 韓 先花, 古川 顕, 金崎 周造

      詳細を見る

      2018年4月 - 2021年3月

      課題番号:18H03267

      配分額:15600000円 ( 直接経費:12000000円 、 間接経費:3600000円 )

      肝臓がんは、平均5年生存率が30%程度とがん全体の中でも極めて低く、正確な診断に基づく早期発見が求められる。肝臓がんの診断には、多時相CT画像(造影剤を急速注入してから経時的に複数回の撮像を行なって得られたCT画像)がよく用いられる。本研究では、肝腫瘍性病変の計算機支援診断法の確立を目的とし、多時相CT画像から有効な時空間特徴抽出のフレームワークを構築するとともに、肝腫瘍性病変の鑑別と診断の参考となる過去の類似肝腫瘍性病変症例を検索できるシステムを開発する。初年度の2018年度において、以下の研究成果が得られた。
      (1)CT画像から高精度で肝臓領域をセグメンテーションする三つの手法(Random WalkとSuper Pixelを組み合わせた手法、深層学習(U-Net)とGANを組み合わせた手法、対話的深層学習法)を開発した。それぞれ国際学術誌CMIG(雑誌発表1), JOIG(雑誌発表2)と国際学会InMed2019(学会発表9, 10)で発表した。
      (2)テンソルスパースコーディング法を開発し、多時相CTの時空間特徴を抽出し、肝腫瘍性病変の鑑別法を開発した。また、近年注目されている深層学習法(CNNとLSTMを組み合わせた手法)を用いた肝腫瘍性病変の鑑別法も開発した。それらの成果は、医用画像分野のトップ国際学会MICCAIなどで発表した(学会発表5, 6, 7)。
      (3)診断の参考となる過去の類似肝腫瘍性病変症例を検索できるプロットタイプシステムを開発し、その有用性を検証した。その成果は国際学術誌JCARS(雑誌発表3)で掲載された。

      researchmap

    • テンソルスパース表現による多時相CT画像の時空間特徴抽出と肝腫瘍性病変の診断支援

      日本学術振興会  基盤研究B 

      陳 延偉

      詳細を見る

      2018年4月 - 2020年3月

      資金種別:競争的資金

      researchmap

    • 人間認知理論と深層統計解析による画像表現と理解

      日本学術振興会  科学研究費助成事業 基盤研究(C) 

      韓 先花, 陳 延偉

      詳細を見る

      2015年4月 - 2018年3月

      課題番号:15K00253

      資金種別:競争的資金

      配分額:4550000円 ( 直接経費:3500000円 、 間接経費:1050000円 )

      本研究では、高信頼性且つ高レベルビジョン段階の画像表現学習を行い、高精度な画像認識・理解に関する研究を進めてきた。主な研究成果は以下の三点である。(1)人間視覚認知プロセスに従って、ロー画像に視覚応答強度ドメインを変換し、全ての情報を保持できる局所記述子を提案した。(2)混交ガウスモデルを用い、トレーニング記述子集合をモデリングし、テスト画像の記述子から高次な統計量を抽出することで、高信頼性な画像表現法を開発した。(3)Middle-levelな画像表現構造を積み重ねることで、深層フレームワークを発展させ、高信頼性且つ高レベルビジョンの画像表現フレームワークを構築した。

      researchmap

    • 変形の伴う3次元形状間の全自動アラインメント

      日本学術振興会  科学研究費助成事業 基盤研究(C) 

      徐 剛, 陳 延偉, 韓 先花

      詳細を見る

      2014年4月 - 2017年3月

      課題番号:26330212

      配分額:4810000円 ( 直接経費:3700000円 、 間接経費:1110000円 )

      医用ボリュームデータにおける肝臓を主成分線形変形(8パラメータ)と位置姿勢スケール(7パラメータ)の計15パラメータに対して全自動で位置決めすることを問題として設定した。全探索というアルゴリズムを考案し、ソフトウェア実装、CUDAによる並列処理による高速化を行った結果、当初設定した問題を完全に解くことができた。従来、ボリュームデータに対して人手による切り出しを行ってきているが、約6時間かかっている作業であるのに対して、本システムでは、約90秒の時間で完成する。GPU以外に高速化を工夫したアルゴリズムは、ピラミッド探索、回転の均等サンプリング、ディスタンスフィールドの事前作成、などである。

      researchmap

    • 変形の伴う3次元形状間の全自動アラインメント

      日本学術振興会  基盤研究C 

      徐剛

      詳細を見る

      2014年4月 - 2016年3月

      資金種別:競争的資金

      researchmap

    • 多次元多様体学習を用いたコア情報の抽出と階層的大規模画像理解の研究開発

      日本学術振興会  科学研究費助成事業 若手研究(B) 

      韓 先花, 陳 延偉

      詳細を見る

      2012年4月 - 2015年3月

      課題番号:24700179

      資金種別:競争的資金

      配分額:3380000円 ( 直接経費:2600000円 、 間接経費:780000円 )

      本研究では、画像を高速かつ高精度で認識するため、外部環境からの影響が受けにくい局所的な特徴を抽出し、統合するため従来のBoFでの量子化ではなく一つのテンソルとして取り扱い、統一的に効率よく記述する枠組を開発した。テンソルになる大量な局所特徴を融合するため多次元教師付き部分空間法を提案し、シーン、物体等への認識応用において大幅な精度の向上を検証した。また画像からhand-craftな局所特徴(SIFTなど)を抽出するだけではなく、簡単且つ効率に得られるraw特徴(色輝度値、勾配など)に対して、データ駆動モデル(k-means、GMM)を用いて解析し、より高速且つ効率的な画像記述法を開発した。

      researchmap

    • 一般化N次元スパースコーディングによる腹部の複数実質臓器統計ボリュームモデリング

      日本学術振興会  科学研究費助成事業 基盤研究(B) 

      陳 延偉, 田中 弘美, 韓 先花, 佐藤 嘉伸, 古川 顕, 森川 茂廣, 健山 智子

      詳細を見る

      2012年4月 - 2015年3月

      課題番号:24300076

      資金種別:競争的資金

      配分額:16250000円 ( 直接経費:12500000円 、 間接経費:3750000円 )

      近年、多くの情報を系統的に集め、情報の簡素化、要約を行う有効な手法としてスパースコーディングという手法が注目されている。本研究では、スパースコーディングを多重線形代数の枠組で拡張させ多次元データ解析に利用できる一般化N次元スパースコーディング法を理論的に開発し、その有効性を示した。また,従来のスパース性に加えて低ランク性を加えたSparse and Low Rank Matrix Decompositionを用いた臓器の局所統計形状モデリング法を提案し,肝硬変症の支援診断に応用し診断精度を95%まで向上させた。

      researchmap

    • ICA縮退フィルタと多重解像度フィルタを用いたIVR画像の画質改善法の開発

      日本学術振興会  若手研究B 

      韓 先 花

      詳細を見る

      2008年4月 - 2011年3月

      資金種別:競争的資金

      researchmap

    • 一般化N次元PCAに基づく人体臓器の統計ボリュームモデリングと診断支援への応用

      日本学術振興会  科学研究費助成事業 基盤研究(B) 

      陳 延偉, 田中 弘美, 田中 覚, 佐藤 嘉伸, 古川 顕, 森川 茂廣, 健山 智子, 韓 先花

      詳細を見る

      2009年 - 2011年

      課題番号:21300070

      配分額:16510000円 ( 直接経費:12700000円 、 間接経費:3810000円 )

      本研究では、多次元データ解析のためのテンソルベース学習法GND-PCA(一般化N次元PCA)を新たに提案した。また、GND-PCAと3次元形状正規化法を用いた臓器の統計ボリュームモデリング法を提案した。3次元形状正規化法は、形状のバリーエションを除去し、濃度値(テキスチャ)のバリーエションのみを表現できるようにするためである。また、GND-PCAは、少数サンプルによる過学習を防ぐ為に用いた。肝臓を対象に構築した統計ボリュームモデルにおいて、少数なサンプルに対してもGND-PCAを用いれば、高い汎化能力をもつ統計モデルを構築することができた。また、構築した統計ボリュームモデルが、正常と異常の識別にもきわめて有用であることを示した。

      researchmap

    • ICA縮退フィルタと多重解像度フィルタを用いたIVR画像の画質改善法の開発

      日本学術振興会  科学研究費助成事業 若手研究(B) 

      韓 先花

      詳細を見る

      2008年 - 2009年

      課題番号:20700170

      配分額:1690000円 ( 直接経費:1300000円 、 間接経費:390000円 )

      本研究では,IVR画像の画質改善および血管強調に焦点を絞り,ICA ShrinkageフィルタとMultiscaleフィルタを組み合わせた適応的(adaptive)な画質改善法の開発により,IVRの画質改善と血管強調の両方を同時に達成できる手法を開発し,その実用化を目指す.研究の成果が以下のとおりです。
      (1) IVR画像における血管信号を強調するMultiscaleフィルタを開発した;
      (2) IVR画像に適したICA Shrinkageフィルタの開発し,量子ノイズを除去した;
      ICA Shrinkageフィルタの結果とMultiscaleフィルタの結果を融合させる方法を開発し,
      画質改善(ノイズ除去)と血管強調が同時にできる手法を確立できた;
      (3) 開発したICA ShrinkageフィルタとMultiscaleフィルタ融合法が実際のIVR動画像を応用した;
      (4) DVBV法を用いた改善画像の定量評価。

      researchmap

    ▼全件表示

    産業財産権

    • 画像解析装置、画像登録装置および画像検索装置

      詳細を見る

      出願番号:2008-151090 

      公開番号:4228031 

      researchmap

    お問い合わせ先
    外部の方
    学内の方