Detection of Skin Cancer Lesions from Digital Images with Image Processing Techniques

Melanoma, a kind of skin cancer, could be a category of cancer that originates from the pigment which includes the cells named melanocytes. Melanomas normally appear within not only the skin, but also may rarely occur within the eyes, intestines, or mouth. In women melanomas most ordinarily found on the legs, on the other hand in men,they are most typical on the rear side. Generally,these are originated from a mole with regarding changes as well as they increase in size, asymmetrical edges, and modification in skin breakdown, itchiness, or color. Melanoma is one of the life-threatening kinds of carcinoma. There is an increase in incidence rates of skin cancer, specifically between non-Hispanic white females and males, however, the chances of survival are high detected in an early stages. In this project, we are taking the help of the image processing techniques for detecting melanoma in the image. Firstly, we will apply the preprocessing technique in order to make the image noise free. Median fillers will be used for filtration of the image. After that, we will transform the image into an HSI color image. Active shape segmentation and texture segmentation techniques will be used for segmenting the image. For feature extraction, GLCM Feature Extraction algorithm is used. Finally,we will apply the probabilistic neural network (PNN) classifier to classify the image either as normal or melanoma.