Skin lesion analysis system for melanoma detection with an effective hair segmentation method

Skin cancer is one of the most challenging disease among various cancer type. According to the studies, there are different types of skin cancers like melanoma, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC). Melanoma is a condition where the production of melanin is significantly reduced because of the dysfunctionality of melanocyte cells. One of the major factor that affects the melanocyte cell is sunburn due to the ultraviolet rays from the sun. The disorder can be identified by the shape, color and texture of skin lesion. The skin lesion can be normally separated into three as benign, atypical and melanoma. A benign skin lesion is a normal skin, atypical skin lesion may or may not be cancerous and melanoma is surely a cancerous one. This paper proposes a non invasive automated skin lesion analysis system for the early detection of melanoma using image processing techniques and mobile technologies. Hair detection and removal is performed for effective classification and extraction features of the skin wound. A fast marching in painting algorithm is used for the hair removal.