Human skin is an exceptional structure. Different kinds of skin cancer exist. And nobody knows about it. Around are many altered types of skin illnesses, some of which are very prevalent. Owing to the dearth of accessible medical facilities, patients in rural places tend to overlook it. Additionally, the diagnosis of this skin cancer is time-consuming. Early recognition of melanoma skin cancer is essential for effective therapy. As of late, it has become widely recognized that melanoma is the most severe type of skin cancer among the others due to its high propensity to spread to other body areas if left untreated. Clinical diagnosis of many disorders is increasingly dependent on non-invasive medical computer vision or medical image processing. These methods offer a computerized analysis of images tool for a quick and precise assessment of the lesion. The research's steps are as follows: gathering the Dermoscopy, or picture database; initial processing; separation via thresholding; mathematical feature extraction through Gray Level Co-occurrence Matrix (GLCM), Asymmetry, Border, Color, Diameter (ABCD), etc.; feature selection via Principal Component Analysis (PCA); calculating the total Dermoscopy Score; and, lastly, classification through convolutional neural network (CNN). A precision of 89.8% and an experimental efficiency of 93.8% were achieved after utilizing the publicly available data set.