Appending Global to Local features for Skin Lesion Classification on Dermoscpic Images
Skin cancer is the most deadliest forms of all other cancers combined; In this paper various pre and post-treatments are presents for improving automated melanomas diagnosis of dermoscopy images. At first pre-processing have done to exclude unwanted parts, a triple A-segmentation proposes to extract lesion according to their histogram patterns. Lastly, suggest appending process with testing many factors for superior detection decision. This paper argues different detection rules: first system used fuzzy rules based on a different features, a second test have been done by modeled local colours with bag-of-features classifier. Then add lesion shape on two previous systems as their global form in the first one, while distributed it and appending with local colour patches in the second system. For each case, different features; various colour models, and many other parameters are examines to decide which settings are more discriminative. Evaluation performance of each method has carried out on (ISIC2019 Challenge) dermoscopic database. The higher classification accuracy results 98.26% with prove its specific parameters that achieved by appending global asymmetric feature to the colour patches.