Fusion of rough set theoretic approximations and FCM for color image segmentation

Abstract

A new technique applying the fusion of rough set theoretic approximations and fuzzy C-means algorithm for color image segmentation is presented. The aim of the technique is to segment natural images with regions having gradual variations in color value. The technique extracts color information regarding the number of segments and the segments center values from the image itself through rough set theoretic approximations and presents it as input to FCM block for the soft evaluation of the segments. The performance of the algorithm has been evaluated on various natural and simulated images.