Border Detection of Skin Lesion Images Based on Fuzzy C-Means Thresholding

摘要

The accurate location of the border of skin lesions is an important first step in the automatic diagnosis of malignant melanoma. In this paper, we propose a new method of segmentation to locate the skin lesion. The method consists to two stages; image pre-processing and image segmentation. As the first step of image analysis, pre-processing techniques are implemented to remove noise and undesired structures for the images using median filtering. In the second step, the fuzzy c-means (FCM) thresholding technique is used to segment and localize the lesion. The border detection results are visually examined by an expert dermatologist and are found to be highly accurate.