Color image segmentation using fuzzy integral and mountain clustering

This paper presents a flexible model for the segmentation of color image data using the fuzzy integral and the mountain clustering. Fuzzy integral is used as a "distance" measure in the mountain clustering applied to find representative regions in the image. The proposed approach does not require an initial estimate of cluster centers for the segmentation process. Segmentation results using the proposed method will depend on the griding of the image space, which specifies the degree of detail in the segmentation process. (C) 1999 Elsevier Science B.V. All right reserved.