The Improvement of Automatic Skin Cancer Detection Algorithm Based on CVQ technique

Abstract

Nowadays, by increasing the number of deaths related to skin cancer, this kind of cancer has been converted as one of the important issues in humans' life. However, the main key is early detection of skin cancer in order to save the life of people. By considering this fact that there is a near similarity between cancer moles and normal ones, attention to artificial systems with the ability of distinguishing between these kinds of moles can be very important, undoubtedly. The accuracy of this kind of system must be considered in order to find better results, especially in the cases which are related to human‘s life. In this paper, with regard to the fact that the raising of a kind of skin cancer, Melanoma, has increasing, we have employed neural networks in the aim of function improvement of an approach based on compressed image technique, namely, Classified Vector Quantization (CVQ) technique. This suggested method has been examined on some images and the results show that this method is a proper way in order to automatic skin cancer detection.