Microaneurysms Detection in Retinal Fundus Images
Sathananthavathi.V1, Indumathi.G2
1Sathananthavathi.V, Department of ECE, Mepco Schlenk Engineering College, Sivakasi (Tamil Nadu), India.
2Indumathi.G, Department of ECE, Mepco schlenk Engineering College, Sivakasi (Tamil Nadu), India.
Manuscript received on 22 April 2019 | Revised Manuscript received on 01 May 2019 | Manuscript Published on 07 May 2019 | PP: 18-23 | Volume-7 Issue-6S3 April 2019 | Retrieval Number: F1004376S19/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© 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: Diabetic retinopathy (DR) is accepted to be the vision threatening disease in most of the developing countries. The anomalies caused by Diabetic retinopathy are identified through Micro aneurysms and Hemorrhages. Manual detection of abnormalities like Micro aneurysm and Hemorrhages in color fundus images is a time consuming process and hence there is a need of an automated system to grade the level of diabetic retinopathy. The proposed method is based on profile based features, local features and SVM classifier to detect microaneurysm in retinal fundus image. The microaneurysm detection accuracy of the proposed method is about 93.73% using SVM classifier and 90.71% for KNN classifier.
Keywords: Fundus Images, Intensity Profile Analysis, Red Lesions, Microaneurysm, Diabetic Retinopathy.
Scope of the Article: Image Security