Automatic detection of Optic Disk and Exudate from retinal images using Clustering algorithm

Diabetic Retinopathy (DR) is a major cause of blindness. Variation in retinal blood vessel thickness, secretion of Exudates which is a protein leakage in the retina, Hemorrhages are some of the symptoms of Diabetic Retinopathy. It is a kind of disorder which occurs due to high blood sugar level. Since Optic Disk appears as a bright spot in the retinal image, which resembles exudates, it has to be removed from the image. Hence detection of Optic Disk is an essential parameter in retinal analysis. In this paper, an automatic and efficient method to detect Optic Disk and exudates are proposed. The real time retinal images are obtained from various eye hospitals. The retinal images are pre-processed using the technique of LAB color space image. The preprocessed color retinal images are segmented using Line operator and Fuzzy C Means Clustering technique in order to detect Optic Disk. The outputs are compared and the best method is determined. The exudates are extracted using K-means clustering and finally the classification is done using SVM.