AUTOMATIC DETECTION AND CATEGORIZATION OF SKIN LESION FOR EARLY DIAGNOSIS OF SKIN CANCER USING YOLO-V3 - DCNN ARCHITECTURE

Abstract

Skin cancer was among the most lethal illnesses in the globe, especially malignant melanoma, benign seems to be a kind of skin cancer that would be the most deadly because of its rapid development, affecting a huge number of individuals globally. Early diagnosis of skin cancer seems to be critical, hence our research employs the YOLO v3 - DCNN architecture to discover and categorize the deadliest kinds of skin cancer. Initially, YOLO v3 generates the feature map, simultaneously color features are extracted using color moments with QuadHistogram, whereas Grey Level Co-occurrence Matrix (GLCM) with Redundant Contourlet Transform(RCT) is generated texture features, and both (color and texture) features are get fused. Then, fused features are fed into the Deep Convolutional Neural Network (DCNN) which classifies the different types of skin cancer. Finally, our proposed approach is compared with the existing works. As a consequence, when contrasted to the baseline techniques, our proposed YOLO-v3 –DCNN has a greater accuracy.