Segmentation of skin cancer images based on gradient vector flow (GVF) snake

Abstract

A gradient vector flow (GVF) snake is proposed in this paper for the segmentation of skin cancer images. In order to make the snake insensitive to noise and be able to remove the hairs, an Adaptive Filter (Wiener and Median filters) is proposed. After the noise and hairs are removed, GVF snake will be used to segment the skin cancer region. The GVF snake extends the single direction and allows it to still be able to track the boundary of the skin cancer even if there are other objects near the skin cancer region. We have proposed new operators to find better edge map in a restored grey scale image. Subjective method has been used by comparing the performance of the proposed gradient vector flow (GVF) snake with other recommended operators of first derivative like Sobel, Prewitt, Roberts and second derivative like Laplacian. The root mean square error and root mean square of signal to noise ratio have been used for objective evaluation. Finally, to validate the efficiency of the filtering schemes different algorithms are proposed and the simulation study has been carried out. Experiments performed on 8(eight) cancer images show the effectiveness of the proposed algorithm.