Robust image segmentation using genetic algorithm with a fuzzy measure

In this paper we present new region-based image segmentation methodology on gray-level images using a genetic algorithm with a fuzzy measure. We first propose a fuzzy validity function which measures a degree of separation and compactness between and within finely segmented regions, and an edge strength along boundaries of all regions. We apply the generic algorithm to search a good or usable region segmentation, which maximizes the quality of regions generated by split- and-merge processing. The iterative algorithm provides a useful method for image segmentation without the need for critical parameters or threshold values, iterative visual interaction or a priori knowledge of an image. Copyright (C) 1996 Pattern Recognition Society.