Lesion detection using segmented structure of retina

Abstract

Morphology of fundus image indicates diseases like diabetic retinopathy and glaucoma. Features of the retinal images allow ophthalmologist to perform retinal disease identification. Presence of lesions in the fundus retinal image is initial sign of diabetic retinopathy. The paper proposes a method for the detection of lesions in retinopathy fundus images based on segmented structure of retina. Morphological operators extract image features and selected features are passed into the support vector machine (SVM) classifier which classifies the images into normal and abnormal classes.