
Retinal Exudates Detection using Binary Operation and Hard Exudates Classification using Support Vector Machine
Author(s) -
Arun Pradeep,
X. Felix Joseph
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.i7572.078919
Subject(s) - support vector machine , artificial intelligence , retinal , computer science , rgb color model , fundus (uterus) , pattern recognition (psychology) , binary number , diabetic retinopathy , computer vision , stage (stratigraphy) , ophthalmology , medicine , mathematics , biology , paleontology , arithmetic , diabetes mellitus , endocrinology
Retinal exudates considered as a symptom of Diabetic retinopathy(DR) is one of most significant reason for visual deficiency. This paper focusses on early detection of hard exudates and to diagnose DR. Binary operations based exudate detection and SVM based hard exudate classification is discussed in this study. The RGB channel of fundus image is converted to HSI colour space for improved noise suppression and optic disc is eliminated preservinsg the blood vessels. In the final stage, hard exudates are classified using SVM classification. In order to evaluate the proposed approach, experiment tests are carried out on different set of images and the results are verified. The results are promising and suggest that the proposed method could be a diagnostic aid for ophthalmologists in the screening of DR