期刊名称：Canadian Journal on Image Processing and Computer Vision
印刷版ISSN：1923-1717
出版年度：2011
卷号：2
期号：2
页码：16-19
出版社：AM Publishers Corporation Canada
摘要：Fuzzy c-means clustering (FCM) with spatial constraints is an effective algorithm suitable for image segmentation. Considering the spatial information of pixels and based on spatially weighted FCM, we propose a new algorithm which improve the accuracy color image segmentation results. Color image features (RGB) are converted into HIS space, and use improved FCM clustering algorithm to get membership function of I (intensity) component of pixels. Finally, a new feature is formed with the membership function values and H (hue) component of pixels. Kmeans algorithm is used for final clustering. The results of experiments show the effectiveness of the proposed method.
