Date of Publication :10th August 2017

Abstract: Diabetic Retinopathy (DR) is an eye disease due to diabetes which causes visual loss. So it is rational to provide treatment for DR at first stages of disease. Hemorrhages are the first symptoms that signify person have diabetic retinopathy. Therefore, their recognition is very essential. In this paper image contrast is improved by pre-processing and then blood vessels are detected as the boundaries of hemorrhages are not differentiate when they are in contact with blood vessels.then classify the image on the basis texture feature such as area, standard deviation etc. finally normal, moderate or Severe DR is detected on the basis of texture feature

Reference :