An automatic based nonlinear diffusion equations scheme for skin lesion segmentation

The clinical diagnosis of skin lesions is not an easy task for dermatologists. A segmentation method that makes the extraction of important characteristics of the skin lesion image can aid dermatologists in the clinical diagnosis of skin cancer. In this paper a segmentation method that combines the use of nonlinear diffusion equations and the Canny edge detector for the automatic detection of the skin edge lesions is presented. Experimental results show the efficacy of the proposed method even when the lesion is noisy or when the lesion is covered by hairs.