ABSTRACT
The processing of images by performing some operations in order to get enhanced images is called as image processing. It is widely used to diagnose the eye diseases in an easy and efficient manner. Several techniques has been developed for the early detection of DR on the basis of features such as blood. It includes the image enhancement processes like histogram equalization and adaptive histogram equalization for the detection of DR. The persistent damage caused to the retina is termed as the retinopathy. The condition of diabetic retinopathy (DR) happens with those who have diabetes that results in progressive damage to the retina. Due to high blood glucose levels it leads to the damage of small blood vessels in the retina and this may result into swelling of the retina. ie., DR is a diabetes related eye disease which occurs when the blood vessels in the retina become swelled and leaks fluid which ultimately leads to vision loss. The DR is regarded as a serious sight threatening condition. The main objective of this method is to detect DR (Diabetic Retinopathy) eye disease using Image Processing techniques. The tool used in this method is MATLAB and it is widely used in image processing. This project proposes a method for Extraction of Blood Vessels from the medical image of human eye-retinal fundus image that can be used in ophthalmology for detecting DR. This method utilizes an approach of Adaptive Histogram Equalization using CLAHE (Contrast Limited Adaptive Histogram Equalization) algorithm with Convolutional Neural Networks algorithm implementation. The result shows that affected DR is detected in fundus image and the DR is not detected in the healthy fundus image and upto 98% of Accuracy can be achieved in the detection of DR Project.
PROJECT OUTPUT
.png)
PROJECT DEMO VIDEO
.png)
