Fuzzy Connectedness Based Segmentation Of Color Image

This dissertation mainly discusses and studies the theories and methods about the large dataset of Chinese Visible Human .On the basis of analyzing the characteristic of domestic color human slice,author proposes an algorithm of tissue segmentation that based on the scaled based vectorial fuzzy connected image segmentation.The main processing problems of the segmentation of the datasets of Visible Chinese Human are large data brocessine,complex interior tissues and blurring at the border of structures.For segmenting the slices effectively, the scaled based vectorial fuzzy connected image segmentation,the algorithm could be applied on the slices of Visible Chinese Human effectively.In this dissertation, author reviews and analyses several kinds of methods of fuzzy connected image segmentation respectively.Several tests have been applied for the segmentation,which qualitively demonstrate the effectiveness of the method,especially for those objects with blurred boundary.This dissertation includes five sections. In the first section, background of visual human is introduced, including the conception, the significance and developed history at home and abroad. In the second section, some image segmentation methods, such as by paralized boundary segmentation,serial boundary segmentatio,by paralized region segmentation and by serial mathematical segmentation,and other methods based on special thories are described. In the third section, the simple method of fuzzy connected image segmentation applied on gray image is set forth, especially the method based on scaled vectorial fuzzy connected image segmentation. In the fourth section, an algorithm based on FCM and fuzzy connectedness was introduced to obtain more acurrate result.Especially a alternation of divergency with Euclid distance is proposed . Furthermore, the result diagrams are given at the end of this section. The whole work of this dissertation is summarized and the consequent work is proposed at last.