Melanoma Detection and Classification in Digital Dermoscopic Images Using Machine Learning K. Senthil Kumar, S. Varalakshmi, G. Sathish Kumar, and T. Kosalai Abstract Melanoma is an exceptional disastrous variety of skin growth. This cate- gory of cancer is disconcerting due to its habit of causing metastasis. Neverthe- less, it sometimes tough to discern it from nevus due to their alike ocular look and symptoms. There are various steps in computer-assisted interpretation systems of melanoma such as image segmentation, edge detection, feature extraction, and clas- sification. Image segmentation on dermoscopic pictures plays an imperative part in discovering melanoma. The existing segmentation methods are K-means clus- tering, convolutional neural network, ROI clustering, histogram equalization, RA pooling method, RGB, HSV, GLCM, Cole–Cole model, multi-variant analysis of variant, fully convolutional neural networks, and mobile imaging system. A support vector machine (SVM) is applied for the analysis of sheath malignancy to iden- tify melanoma or nevus. The goal is to evaluate the effectiveness of the proposed segmentation method, highlights the most appropriate options, and compares the classification methods. Keywords Melanoma · Skin cancer · Malignant detection · Dermoscopic images · Image segmentation · Hue value saturation · Classification K. S. Kumar (B ) Rajalakshmi Institute of Technology, Chennai, Tamil Nadu, India e-mail: senthilkumar.k@ritchennai.edu.in S. Varalakshmi Aadhi College of Engineering and Technology, Tamilnadu, India G. S. Kumar SCSVMV University, Kanchipuram, Tamilnadu, India T. Kosalai University College of Engineering Kancheepuram, Tamilnadu, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 V. Suma et al. (eds.), Inventive Systems and Control, Lecture Notes in Networks and Systems 204, https://doi.org/10.1007/978-981-16-1395-1_36 493