Skin lesion segmentation using the D-S evidence theory based on the FCM using feature parameters

Lesion segmentation is an important step in diagnostic of skin lesion images. In this paper, to improve the segmentation of the skin color image (RGB), we apply the Fuzzy-C-Means method based on feature parameters in each color component (R, G and B) and we use the Dempster-Shafer evidence theory method to merge the results and to have the final segmentation result. In this work, the proposed technique was compared with the FCM based on feature applying in gray level skin image and the experiments show the performance of the proposed technique.