Abstract
Diabetic Retinopathy is a progressive eye diseases that causes changes in the blood vessels of the retina which may cause blindness if not prevented and treated at early stage. Diabetic retinopathy is one of the complications caused by diabetes and it appears in the retina, which is the tissue responsible for the vision in the eye. The early detection and diagnosis is essential to save the vision of diabetic patients. The indications of diabetic retinopathy on the surface of the retina are microaneurysms, hemorrhages and exudates. The identification of exudates by opthalmolgists normally requires dilation of pupil's eye using a chemical solution which consumes lot of time and affects patients. The presented work investigates and presents a morphology based techniques for the detection of diabetic retinopathy through exudates from color fundus images which includes the elimination of optic disc and the detected exudates are classified using image processing methods.