Pyramid segmentation of color images using fuzzy c-means clustering algorithm

An efficient segmentation method based upon fuzzy c-means (FCM) clustering is proposed. The approach utilizes a pyramid data structure for the hierarchical analysis of a color image, the FCM, and a modified fuzzy c-means (MFCM) clustering algorithm for the segmentation in each level of the pyramid. The feasibility of the method is evaluated using a color image. The results show that the computation time with the proposed method is about one-thirteenth of that which is needed when using the FCM algorithm without pyramid data structure, while the segmentation results are almost the same.<>