Abstract
A greater proportion of people are affected by skin cancer, particularly melanoma, which has a higher tendency to metastasize. For Dermatologist, Visual inspections are most challenging & complex task for melanoma detection. To solve this problem, dermoscopic images are analyzed and segmented. Due to the sensitivity involved in surgical operations, existing techniques are unable to achieve higher accuracy. As a result, computer-aided systems are essential to detect & segment dermoscopic images.
 In this paper, for segmentation 5000 skin images were taken from the HAM10000 dataset. Prior to segmentation, preprocessing is done by resizing images. A novel U Net structure is a fully convolutional network is presented & implemented using up-sampling and down-sampling technique with Rectified Linear Units (ReLU) for activation functions. The outcomes of proposed methodology shows performance improvement for skin-lesion segmentation with 94.7 % pixel accuracy & 89.2 % dice coefficient compared with existing KNN & SVM techniques.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal on Recent and Innovation Trends in Computing and Communication
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
