EARLY STAGE DETECTION OF DIABETES FROM FUNDUS IMAGE OF EYE USING DIABETIC RETIONOPATHY

Diabetes is a quickly expanding overall issue. The most widely recognized confusion of diabetes is diabetic retinopathy (DR), which is one of the essential drivers of visual deficiency and visual hindrance in grown-ups. In this paper, a calculation is proposed for the early and exact detection of the diabetic retinopathy. Fundus images are taken from DRIVE database. In the initial stage image undergo pre-processing steps. Canny edge detection is also applied to extract the fragile blood vessels. Performance of the algorithm is calculated by using accuracy, AUC and specificity.