TY - JOUR ID - 87402 TI - Deep Learning Based Skin Lesion Segmentation and Classification of Melanoma Using Support Vector Machine (SVM) JO - Asian Pacific Journal of Cancer Prevention JA - APJCP LA - en SN - 1513-7368 AU - R D, Seeja AU - A, Suresh AD - Research Scholar, Department of Computer Science, Periyar University, Tamil Nadu, India. AD - Principal, Siri PSG Arts and Science College for Women, Sankagiri, 637301, Salem, Tamil Nadu, India. Y1 - 2019 PY - 2019 VL - 20 IS - 5 SP - 1555 EP - 1561 KW - Melanoma KW - Deep Learning KW - Dermoscopy KW - Segmentation KW - Classification DO - 10.31557/APJCP.2019.20.5.1555 N2 - Objective: The main objective of this study is to improve the classification performance of melanoma using deeplearning based automatic skin lesion segmentation. It can be assist medical experts on early diagnosis of melanomaon dermoscopy images. Methods: First A Convolutional Neural Network (CNN) based U-net algorithm is used forsegmentation process. Then extract color, texture and shape features from the segmented image using Local BinaryPattern ( LBP), Edge Histogram (EH), Histogram of Oriented Gradients (HOG) and Gabor method. Finally all thefeatures extracted from these methods were fed into the Support Vector Machine (SVM), Random Forest (RF), K-NearestNeighbor (KNN) and Naïve Bayes (NB) classifiers to diagnose the skin image which is either melanoma or benignlesions. Results: Experimental results show the effectiveness of the proposed method. The Dice co-efficiency valueof 77.5% is achieved for image segmentation and SVM classifier produced 85.19% of accuracy. Conclusion: In deeplearning environment, U-Net segmentation algorithm is found to be the best method for segmentation and it helps toimprove the classification performance. UR - https://journal.waocp.org/article_87402.html L1 - https://journal.waocp.org/article_87402_98260e432395c522b6fee7fdd22ab3e5.pdf ER -