Malignant melanoma is the deadliest form of all skin cancers. This cancer develops when an unrepaired DNA damage is caused to the skin cells which in-turn makes the skin cells to increase rapidly and form malignant tumours. These tumours originate in the pigment-producing melanocytes in the basal layer of the epidermis. Its diagnosis is crucial, if not detected in an early stage. The proposed work comprises of three steps: In the first step, the lesion region is segmented using edge detection

more »

... sed segmentation. In the second step, different types of color, border and texture features are extracted from dermoscopic images based on distinguished structures. In the third step, extracted features are fed to the classifier to classify melanoma using Radial Basis Function classifier. These three steps are used to differentiate the benign tumours from malignant categories. Qualitatitive and quantitative analysis are analysed. Sensitivity, Specificity and Accuracy parameters are greatly enhanced by the use of the improved algorithms.