Skip to Main Content
Diabetic Retinopathy (DR) is a complication of diabetes which leads to vision deterioration and causes total blindness in diabetic patients. Exudates are one of the most prevalent earliest clinical signs of retinopathy. Thus, earlier identification and classification of exudates from retinal images is clinically important which facilitates the ophthalmologists in accurate diagnosis and treatment planning. The images from the hospitals were taken as reference for this purpose. The proposed system is based on automatic diagnosis of retinal diseases such as DR and main goal is to extract features that describe the homogeneity of localized areas (exudates) of the retinal images. The retinal diseases affects the features of eye in a unique manner and extracted feature set helps in distinguishing four diseases (in this research) by the classifiers. Besides this, the technique is database independent since the features are specifically tuned to the pathologies of human eye.