Description

The eye is vital organ and it is the main organ that is mainly affected due to diabetes. Different types of diseases were commonly present in eye. The detection of exudates and hemorrhages in retinal images will be helpful in the early identification of Diabetic Retinopathy and it can helpful in providing treatment for patients. The images are first identified to be normal or exudates affected or hemorrhage affected. The identified defects were then segmented from the images. The performance of the process is measured by measuring the classification parameters such as Accuracy, Sensitivity and Specificity. The retinal images were initially resized and they are filtered using median filter. The filtered images is then classified into Normal, Hemorrhage affected or exudates affected. The GLCM features were used for the classification of the retinal images. Different segmentation methodologies were employed for different types of defects identified. The exudates were segmented using morphological operations. The hemorrhages were segmented using thresholding process. The performance of the process is measured by calculating the performance metrics of the classifier.