• AJAYDILIPKUMAR MARAPATLA
• L DIVYA
• KRISHNA TEJA JILLELAMUDI
• SHUVETHRAM RAGUPATHY

Skin disease is the most widely recognized malignant growth all around the world. including most of skin disease cases with melanoma being the deadliest structure. In the statistic shading pictures of skin, there is high similarity when various skin injury like melanoma and nevus are considered, these increment the trouble of discovery and determination of the skin disease. A dependable computerized framework for skin injury order is basic for early discovery to spare exertion, time and most importantly, human life. In this paper, an automatic skin sore arrangement strategy is proposed. This article depicts the structure, usage, and results of the most recent portion of the dermoscopic picture investigation benchmark challenge. The objective is to help explore what's more, advancement of calculations for computerized determination of melanoma, the deadliest skin malignancy. The technique under consideration here comprise of a pre- trained deep learning system and transfer learning. Notwithstanding calibrating and information enlargement, the exchange learning is connected to a pre-trained model called MobileNet by supplanting the last layer by a softmax to characterize nine distinct injuries (Actinic keratosis, Basal cell carcinoma, Benign keratosis, Dermatofibroma, Melanocytic nevus, Melanoma, Squamous cell carcinoma, Vascular sore, none of the others). The proposed model is prepared and tried utilizing the ISIC 2019 data set. The notable quantitative measures like precision, top_2_accuracy and top_3_accuracy are utilized in assessing the exhibition of the proposed technique where the acquired estimations of the semeasures are 77.70%,92.02% and 98.93% separately. The exhibition of the proposed technique is contrasted and the current strategies where the order rate of the previous beat the presentation of the last strategies.

Dermatology, Dermoscopy, melanoma, skin cancer, lesion segmentation, Disease classification, feature detection algorithm, deep learning, transfer learning, MobileNet, softmax.

"SKIN LESION ANALYSIS TOWARDS MELANOMA DETECTION WITH DEEP NEURAL NETWORKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.723-727, June-2019, Available :http://www.jetir.org/papers/JETIR1906E58.pdf

"SKIN LESION ANALYSIS TOWARDS MELANOMA DETECTION WITH DEEP NEURAL NETWORKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp723-727, June-2019, Available at : http://www.jetir.org/papers/JETIR1906E58.pdf