Skin Cancer Detection Using Gray Level Co-occurrence Matrix Feature Processing

At present time, skin cancer is becoming common explanation for death in citizenry . Often when body exposed to the daylight , it's going to causes carcinoma it's a abnormal growth of skin cells within the physical body . Generally most of the skin cancers are often cured if they're detected in early of its stage. Hence if it's detected early and fast the lifetime of patient are often saved. General method for diagnosis of carcinoma is biopsy. In biopsy affected somatic cell are removed which sample are sent for laboratory testing. it's tedious and time consuming process. So there's a requirement of auto software aided system for accurate and fast processing. it'll empower target understanding by making utilization of quantitative parameters. during this system features of cancerous region are extracted and support vector machine (SVM) classifier is employed to detect carcinoma . This diagnosing methodology uses the pictures taken by dermoscopy, then some image preprocessing is completed to reinforce the standard and take away the noise from images followed by segmentation using thresholding technique. To extract the features of image GLCM methodology is implemented, these features are given as an input to the classifier. Classifier will categories the given image into either of cancerous or non-cancerous type accordingly. The performance analysis indicates that this method outperforms as compared to the prevailing systems as its accuracy is 94.05%.