When a person has diabetic retinopathy (DR), even after having the condition for a long time, they are highly unlikely to be aware of it. Not everyone is really familiar with this illness. This illness is a little different from others since, depending on the diagnostic syndrome, every diabetes patient has a risk of developing diabetic retinopathy. Various studies have been done in this area, but a good method is still needed. A neural network in machine learning needs to be trained very well because if it isn't, the system won't be able to provide decent results. The rate of false alarms is higher due to poor training. However, there is another method—an edge detection tool—by which DR may be detected more accurately. Edge hasthe ability to extract the geometry of impairments, and the density of the retrieved region determines whether or not it is diabetic retinopathy. The exudates from the fundus picture are extracted by the proposed method using the Sobel Edge Detection tool. Prior to that approach, a colour mapping tool was used to make exudates from the fundus picture more visible. A colour mapping tool helps improve the visibility of some patches that the illness may cause. The backdrop can also be classified by changing the colours such that exudates are more obvious than in the original image. The suggested system has more accuracy than the existing model and is tested using the Messidor benchmark.

Keywords : Diabetic Retinopathy, Fundus Image, Sobel EdgeDetection, Color Mapping, Retina, Optic Cup.