An Approach for Diabetic Retinopathy Detection using Image Processing Techniques
Amruta V. Anasane

Amruta V. Anasane, PG Student, Department of Computer Science and Engineering, Prof Ram Meghe College of Engineering & Management, Badnera-Amravati (Maharashtra)-444701, India.
Manuscript received on January 12, 2018. | Revised Manuscript received on January 15, 2018. | Manuscript published on March 30, 2018. | PP: 5-9 | Volume-8 Issue-1, March 2018. | Retrieval Number: A3109038118/2018©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: The medical system for computer-aided screening and grading of diabetic retinopathy which is depended on reliable detection of retinal lesions in infected images. The automatic detection of both microaneurysms and hemorrhages in color fundus images is described and validated in this paper. The various such as Gaussian Mixture Model, the k-nearest neighbor and support vector machine (SVM), AdaBoost are accepted to be analyzed for classifying retinopathy lesions from non-lesions. SVM classifiers are found to be the classifiers for bright and red lesion categorization. This lesion classification problem deals with unstable data sets and SVM or mixture classifiers derived from SVM using more classification error due to the data imbalance. Diabetic retinopathy is a difficulty of diabetes that can lead to impairment of vision and even blindness. It is the most common cause of blindness in the working age population. DR can be manage using available treatments, which are effective if diagnosed early. Since DR is asymptomatic until late in the disease process, regular eye fundus examination is necessary to check any changes in the retina. The several ways in which image analysis helps to examine DR from colour fundus images of the human retina. In order to give contribution, the image enhancement, by which the contrast and the sharpness of the images are enhanced to reduce noise.
Keywords: Fundus images, SVM, Diabetic retinopathy (DR), micro aneurysms (MA) and hemorrhages (HE) etc.