A Novel Technique on Detect Melanoma in Dermoscopy Images By using Deep Learning
Tarun Dhar Diwan1, Upasana Sinha2, Siddhartha Choubey3
1Tarun Dhar Diwan, M.Tech, Computer Science and Engineering, CSVTU, Chhattisgarh.
2Dr.Upasana Sinha, Associate Prof., Department Compter Science & Engg., J.K. Institute of Engineering, Bilaspur.
3Dr. Siddhartha Choubey, Associate Prof., Department , Compter Science Engg.,Shri Shankaracharya College of Engineering, Bhilai.
Manuscript received on December 14, 2019. | Revised Manuscript received on December 24, 2019. | Manuscript published on January 10, 2020. | PP: 1645-1648 | Volume-9 Issue-3, January 2020. | Retrieval Number: C8533019320/2020©BEIESP | DOI: 10.35940/ijitee.C8533.019320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Melanoma is a typical sort of malignant growth that influences countless. As of late, profound learning strategies have been appeared to be very precise in arranging pictures in different fields. This investigation utilizes profound figuring out how to consequently distinguish melanomas in dermoscopy pictures. To begin with, we preprocess the pictures to evacuate undesirable antiques, for example, hair, and afterward consequently fragment the skin sore. We at that point group the pictures utilizing a convolution neural system. To assess its viability, we test this classifier utilizing both preprocessed and natural pictures from the PH2 dataset. The outcomes a remarkable execution as far as affectability, explicitness, and exactness. Specifically, our methodology was 93% exact in distinguishing the nearness or nonappearance of melanoma, with sensitivities and specificities in the 86%– 94% territory.
Keywords: Image Processing, Melanoma Detection, Deep learning, Dermoscopy Image.
Scope of the Article: Deep learning