Skin disease detection and classification

Skin diseases are most commonly occurring in people of all ages and are caused by bacteria, infection or radiation. These diseases have various dangerous effects on the skin and keep on spreading over time. A patient can recover from skin diseases if it is detected and treated in the early stages and this can achieve cure ratios of over 95%. Hence, it is important to identify these diseases at their initial stage to control them from spreading. Skin diseases are primarily diagnosed visually, beginning with an initial clinical screening and followed potentially by dermoscopic analysis. Such a system is often prone to errors. The main idea of this project is to improve the accuracy of diagnostic systems by using Image Processing and classification techniques. In the proposed system, an image captured on camera is taken as input. This image will be pre-processed in order to make it suitable for segmentation by using Contrast Enhancement and Grayscale Conversion. Global Thresholding technique is used to segment the pre-processed image through which the actual affected region is obtained. Texture features, such as Energy, Entropy, Contrast, IDM, are extracted from the segmented image using Grey Level Co-occurrence Matrix. Image Quality Assessment features such as MSE and PSNR are extracted. The extracted texture features will be used to detect the presence of skin disease and classify the disease as melanoma, leprosy or eczema, if present, using the Decision tree technique.