Identification and Enhancement of Diabetic Retinopathy Images Using Filtration Techniques

Abstract

Diabetic Retinopathy is an eye disease that is vision threatening and caused due to high blood sugars due to diabetes. This disease is further classified in various stages and a person may be blind forever if not detected at an early stage. An early detection of this disease is required but due to the high number of patients and very high cost involved in various tests done by ophthalmologists only very few patients report timely. There is a need of an automated method which may be used for early detection of diabetic retinopathy for accurate and precise diagnosis of the disease. In this paper, an approach for comparing the various filtration techniques has been proposed so as to better detect any type of fundus in the eye. The performance is tested with fundus images from different severity levels of Diabetic retinopathy. The efficacy of proposed work has been quantitatively measured by entropy and standard deviation. The resultant images will provide enhanced images for doctors to analyze and may help in detection of diabetic retinopathy more accurately at early stages