Dermoscopy Image Segmentation Using a Modified Level Set Algorithm

Border segmentation serves as the first step in the automated analysis of dermoscopy images for the purpose of diagnosing melanoma and other pigmented skin lesions. Increasing the efficiency of the level set method for the purpose of an appropriate segmentation is in direct relationship with initialization and estimating the level set controlling parameters. In this paper, using a modified level set algorithm, an inventive method for the segmentation of skin lesions is proposed. This algorithm can directly evolve from the initial segmentation by creating a weighted combination of different segmentation methods such as Fuzzy C-means (FCM) clustering, k-means clustering, Otsu and Iterative Methods. In addition, it will be possible to estimate the controlling parameters of the level set evolution from the obtained initial contour. The results of the proposed method and comparison with the labeled segmentation results by specialists have been obtained on a standard the dermoscopy image database.