Segmentation of Exudates from Non Exudates in Color Fundus Images

Diabetic retinopathy is a major cause of preventable blindness in the world. Regular screening is essential in order detect the early stages of diabetic retinopathy for timely treatment to prevent or delay further deterioration. This project detects the presence of abnormalities in the retina such as the exudates in retinopathy images using computational intelligence techniques. The color retinal images are segmented using fuzzy c-means clustering following color normalization and contrast enhancement. For classification of these segmented regions into exudates and nonexudates, a set of initial features such as color, size, edge strength, and texture are extracted. A genetic based algorithm is used to rank the features and identify the subset that gives the best classification results. Using a multilayer neural network classifier, the selected feature vectors are then classified.