Skin Cancer Image Detection using Watershed Marker-Controlled and Canny Edge Detection Techniques

Norhafiza Hamzah, Sapaweh Asli, Rechard Lee
2018 Transactions on Science and Technology   unpublished
There are two main types of skin cancer which are melanoma and non-melanoma. This paper proposed a prototype system for detecting melanoma type of skin cancer. The detection system developed in this paper is based on digital images of skin cancer using a combination of watershed marker-controlled technique and canny edge detection technique during image segmentation process. The ABCD (Asymmetry, Border, Colour, and Diameter) method is used to extract the characteristics of the melanoma type of
more » ... kin cancer on the skin cancer image. The four characteristics of melanoma skin cancer are Asymmetrical shape, Border irregular, Colour variant and Diameter greater than 6mm, which are used by dermatologists to classify melanomas. Several experiments had been conducted to test the accuracy and effectiveness of the proposed system. The experiments are detection test for melanoma skin cancer melanoma and non-melanoma skin cancer image, comparison test for image segmentation techniques, and a detection test using the noised image of melanoma skin cancer. The results show the proposed prototype system is able to detect nine out of ten images of melanoma skin cancer and distinguish eight out of ten for non-melanoma skin cancer images. Combination of the proposed techniques is better compared to the other techniques because of the segmentation techniques used can distinguish between the main object, Region of Interest, as well as the less important object in the image of skin cancer.