Skin cancer exists in different forms such as: Melanoma, Basal and Squamous Cell Carcinoma, among which Melanoma is the most dangerous and unpredictable skin cancer. In this paper, we implement an image processing technique for the detection of Melanoma skin cancer, in which the input to the system is the skin lesion image. This image proceeds with the image preprocessing methods such as conversion of (Red, Green, Blue)RGB image to gray scale image, noise removal, hair removal and so on. Moreover thresholding is used to segment the images followed by feature extraction that includes parameters such as: Asymmetry, Border irregularity, Color and Diameter (ABCD). Finally, the total Dermoscopic score (TDS) is calculated and decision is to be taken whether the cancer is Melanoma or not