Skin lesion segmentation using co-operative neural network edge detection and colour normalisation

Lesion segmentation is an important step in analysing dermoscopic skin lesion images. In this paper we achieve accurate lesion segmentation using a co-operative neural network-based edge detection approach coupled with a pre-processing step that enhances colour information and contrast of the images. Extensive experiments are carried out on a dataset of 100 dermoscopic images with known ground truths obtained from three expert dermatologists. The results show that our approach is capable of providing good segmentation performance while the colour enhancement step leads to improved segmentation accuracy.