Hair Removal Combining Saliency, Shape and Color

In a computer-aided system for skin cancer diagnosis, hair removal is one of the main challenges to face before applying process automatic lesion segmentation and classification. this paper, we propose straightforward method detect remove from dermoscopic images. Preliminarily, regions consider as candidate border/corner components located on image frame are automatically detected. Then, determined using information regarding saliency, shape colors. Finally, detected restored by simple inpainting method. The evaluated publicly available dataset, comprising 340 images in total, extracted two commonly used public databases, an specific dataset including 13 already other authors evaluation comparison purposes. We also qualitative quantitative results promising detection accurate, performance satisfactory existing methods.