题目：Extension of FCM for Clustering, Image Segmentation and Motion Detection Application (模糊c均值在聚类，图像分割和运动检测中的应用)

 

摘要：Fuzzy c-means (FCMs) have been considered as an effective algorithm for pattern recognition, image segmentation and motion detection. In this talk, we introduce how to modify objective function in FCMs for various different applications. First, we introduce a more flexibility function which considers the distance function itself as a sub-FCM. The sub-FCM distance function in HFCM is general and flexible enough to deal with non-Euclidean data. Second, we impose generalized mean on membership to incorporate local spatial information and cluster information, and on distance function to incorporate local spatial information and image intensity value. Thus, our GFCM is more robust to image noise with the spatial constraints: the generalized mean. Finally, to extend FCM for 3D video analysis, we present a dynamic fuzzy clustering for automatically detecting time varying characteristics and phenomena.

 

张辉，分别于2003，2006和2010年于东南大学获得无线电工程系学士，生物医学工程系硕士和计算机科学与技术系博士学位。2010年至2013年在加拿大温莎大学担任Research Associate。现为南京信息工程大学教授，图像处理实验室负责人。张辉教授的研究方向为模式识别，图像处理，计算机视觉等。目前已在国际权威杂志IEEE Transaction on Image Processing，IEEE Transaction on Neural Networks and Learning Systems，IEEE Transaction on Geoscience and Remote Sensing等发表SCI论文二十余篇，以及国际著名会议ICIP，ICPR和ICASSP等发表EI论文十余篇，论文的引用率已超过300次。张辉教授是IEEE Trans. Circuits and Systems for Video Technology，IEEE Trans. Fuzzy System，IEEE Trans. Information Forensics & Security，IEEE Trans. Neural Networks and Learning Systems，IEEE Trans. Systems, Man, and Cybernetics, Part B: Cybernetics，Pattern Recognition，Digital Signal Processing，Signal processing，Neurocomputing等杂志的审稿人。