Hair Removal and Lesion Segmentation with FCN8-ResNetC and Image Processing in Images of Skin Cancer

Skin cancer is quite common. Early detection crucial for the treatment of skin cancer. Methods based on computer technology (deep learning, image processing) are now increasingly used to diagnose These methods can eliminate human error in diagnostic process. Removing hair noise from lesion images essential accurate segmentation. A correctly segmented increases success rate diagnosing In this study, a new FCN8-based approach removal and segmentation presented. Higher was achieved by adding ResNetC FCN8. model ResNet. Two datasets were study: ISIC 2018 PH2. Training 89.380% 97.050% 3000 masks created as part study remove images.