Review of segmentation methods on malignant melanoma

Dermoscopy is a non-invasive, widely used diagnostic tool that aids the diagnosis of skin lesions and is proven to increase the accuracy of melanoma diagnosis. If the melanoma is diagnosed in an early stage, there is a high probability of being cured. In this regard, several imaging techniques have been explored to improve the diagnosis accuracy of skin lesions. The first step in these systems is skin lesion segmentation. The next essential step is feature extraction and pattern analysis procedures for better recognition. In this paper, a brief study has been carried out about the existing segmentation methods adopted by different scholars such as Adaptive thresholding, Thresholding based on type-2 fuzzy logic, Statistical region merging, Neuro fuzzy model, Geometric deformable models, Levelset model, Wavelet network model. The segmentation methods discussed here are thresholding, edge-based, and region-based. Color images when segmented directly, yields better differentiation between the lesions. These techniques are evaluated using Accuracy rate, sensitivity, specificity, Border error, Hammoude distance, Hausdorff distance, MSE, PSNR and elapsed time.