Skin cancer has recently become one of the types of cancer that often appears. In Indonesia, skin cancer is placed third after cervical and breast cancer. Mortality from skin cancer could be reduced if the detection and treatment are early and appropriate. The objective to be achieved in this research is to give a better model with a non-invasive method to classify the skin lesions of melanoma using Support Vector Machine.
The framework of the research can be divided into five parts: (1) Preparation of the dataset that is used as data input (Dataset was provided by ISIC The International Skin Imaging Collaboration) (2) Pre-Processing is the process of removing hair components that are still present in the dermoscopy (3) Image Segmentation, searching for the edges of the lesion (4) Feature Extraction, parameters to distinguish the two types of lesions, and (5) Classification, the classification process using the support vector machine method of skin lesions whether they are melanocytes or non-melanocytes.