An Automatic Segmentation on The Dermoscopy Images for Organizing Uneven Borders Using Iterative Contours
Year: 2025
Type: article
Abstract: Among all malignancies, skin disease is the most overwhelming and destructive disease. This ailment begins first from the epidermis of the human body. To acquire a precise assessment of skin malignant growth, the electronic examination of the image produces a profound impact. All around the globe, skin disease affects numerous individuals in various parts of the body. To make an ideal conclusion of skin disease, the dermatologist ought to look at the color and border of the skin picture using a computational technique. This could be a pre-screening framework for the dermatologist for an early conclusion. The proposed work reports on the division of sore from the dermoscopy pictures with the essential steps, for example, pre-processing, segmentation, and post-processing. The Region Growing and dynamic shape identification makes an ideal bend as a limit to the portion influenced district. The proposed work starts with separating, followed by division and closures with include extraction, and this is clarified impeccably. The proposed reproduction gauges the exact finding between the Ground Truth image and the Segmented Image and affirms the best-offered estimations of exactness up to 93.43% for the DermQuest dataset and 92.74% for the DermIS Dataset.
DOI: 10.14419/8zw85s25
Authors: A. Prabhu Chakkaravarthy, Prof. Chin-Shiuh Shieh, Prof. Mong-Fong Horng, J Dhanalakshmi
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4413220397