MULTIRESOLUTION COLOR IMAGE SEGMENTATION

Image segmentation is the process by which an original image is partit ioned into some homogeneous regions. In this paper, a novel multiresol ution color image segmentation (MCIS) algorithm which uses Markov Rand om Fields (MRF's) is proposed. The proposed approach is a relaxation p rocess that converges to the MAP (maximum a posteriori) estimate of th e segmentation. The quadtree structure is used to implement the multir esolution framework, and the simulated annealing technique is employed to control the splitting and merging of nodes so as to minimize an en ergy function and therefore, maximize the MAP estimate. The multiresol ution scheme enables the use of different dissimilarity measures at di fferent resolution levels. Consequently, the proposed algorithm is noi se resistant. Since the global clustering information of the image is required in the proposed approach, the scale space filter (SSF) is emp loyed as the first step. The multiresolution approach is used to refin e the segmentation. Experimental results of both the synthesized and r eal images are very encouraging. In order to evaluate experimental res ults of both synthesized images and real images quantitatively, a new evaluation criterion is proposed and developed.