Main Article Content
Abstract
-Diabetic Retinopathy (DR) is a disease which is caused by high blood sugar from diabetes. It is associated with damage to the tiny blood vessels in the retina. It causes blood vessels in the retina to leak fluid or hemorrhage, distorting vision. The retina detects light and sent to the brain as a signal through the optic nerve. The early stages of diabetic retinopathy have no symptoms. The disease is unnoticed until it finally affects the vision, even though it is often progressed. In its advanced stage, there are new abnormal blood vessels proliferate on the surface of the retina, which leads to scarring and cell loss in the retina. Floating spots appear because of bleeding from abnormal blood vessels. The spots may sometimes clear on their own, but without prompt treatment, Bleeding often recurs and increase the risk of permanent vision loss. Treatment interventions at early stages of diabetic retinopathy can reduce burden of blindness. During the imaging process, artefacts such as eyelashes and eyelids are also imaged along with the retinal area. This brings a big challenge on how to exclude these artefacts.
Excluding the artefacts from true retinal area is important pre-processing step before detection of eye diseases. In computer vision applications Super pixels are becoming increasingly popular because distinguishing the artefacts from true retinal area in SLO images is challenging task. The SLO (Scanning Laser Ophthalmoscope) provides an image of width up to200 degree. In this Paper, a Simple Linear
Iterative Clustering (SLIC) is used which is a super pixel generation method used to avoid data redundancy and it was proved to be efficient in terms of region compactness, computational time. Super pixels reduce the computing cost because it is used to represent different irregular regions in a compact way .A classifier has been built Based on selected features to extract out true retinal area. The experimental evaluation results have shown good performance with an overall accuracy of 93%.