Skin lesion image segmentation using a color genetic algorithm

Abstract

The development of computer-aided diagnosis systems for skin cancer detection has attracted a lot of interest in the research community. In particular, the availability of an accurate automatic segmentation tool for detecting skin lesions from background skin is of primary importance for the overall diagnosis system. In this paper we investigate the capability of a color image segmentation method based on Genetic Algorithms in discriminating skin lesions. Experimental results show that the segmentation approach is able to detect lesion borders quite accurately, thus coupled with a merging technique of the surrounding region could reveal a promising method for isolating skin tumor.