Abstract
Malignant Melanoma is considered as the one of the most deadly skin disease if ignored without treatment. The mortality rate caused by melanoma is more than tow times in comparing of other skin malignancies disease. These facts encourage computer scientist to find automated methods to discover the skin cancers. Nowadays, the analysis of skin images is widely used for assistant physicians to discover the first stage of the disease automatically. One of the challenge the computer science researchers faced when try to develop such system is un-clarity of the exist images such appearing of noise like shadows, low contrast, hairs and specular reflections which complicates the process of detect the skin lesions in that images. In this paper, the solution of the above mentioned problem has proposed by using the active contour method, but active contour has a major drawback of seed selection where it should start to process segmentation. In this paper, fuzzy entropy based morphological processing method has been used to find out automatic seed point for active contour. By incorporating this, it can segment the lesion from dermoscopic images automatically. The proposed methodology was tested on standard dataset DermIS and both quantitative as well as qualitative measures was used to check the reliability of the proposed method that shows promising results.