Abstract: Hyperglycemia and diabetes result in vascular complications, most particularly diabetic retinopathy (DR). The prevalence of DR is increasing and is a most important cause of blindness and visual impairment in developed countries. Current methods of detecting, screening, and monitoring DR are based on subjective human evaluation, which is also slow and time-consuming. As a result, initiation and progress monitoring of DR is clinically hard. Computer vision methods are developed to separate and detect two of the most common DR functions—dot hemorrhages (DH) and exudates. Diabetic retinopathy, an eye disorder caused by diabetes, is the main cause of blindness. This may result in an unprecedented number of persons becoming blind unless diabetic retinopathy can be detected early. Hence here we are trying to detect all potential exudate regions in a fundus image of format .png. And the algorithm counts the number of pixels present in the area of potential Exudates. And also we are trying to detect the position of all size Dot hemorrhages and count the same. And the algorithm places circle around all the detected smaller and larger size Dot hemorrhages to visualize the presence of smaller and larger size Dot hemorrhages.

Keywords: Fundus Images, Hemorrhages, Exudate Image Processing, Diabetic Retinopathy.