Effective Segmentation of Iris in Noisy Eye Images Using C-means based on Grasshopper Optimization Algorithm

Please enable javascript in your browser.

Abdulkhaleq Abd Oun, Mazin | 2021

214 Viewed

Type of Document: M.Sc. Thesis
Language: English
Document No: 54513 (52)
University: Sharif University of Technology
Department: Materials Science and Engineering
Advisor(s): Peyvandi, Hossein
Abstract:
Iris segmentation is an essential step in a medical imaging-based automated diagnosis system. The operation of dividing a digital image into areas with different characteristics and setting objectives is known as image segmentation. The extraction of the iris from unnecessary sections of the image is very important for any biometric device. Eyelid / eyelash obstruction, special reflections, intensity heterogeneity, and an irregular iris border, are considered as noise which affect the iris segmentation process. Focusing on Fuzzy C-Means (FCM) clustering and Grasshopper Optimization Algorithm (GOA), we present a new and efficient method for dividing iris-filled, annoying, and iris limits that aren't circular. The method proposed is benchmarked on the database it is MMU1 to test and verify its performance qualitatively
Keywords:
Digital Images ; Optimization ; Grasshopper Optimization Algorithm (GOA) ; Iris Segmentation ; Fuzzy C-Means (FCM)Clustering ; Medical Imaging

No TOC