Vol 4 , Issue 1 , January - March 2016 | Pages: 51-56 | Research Paper
Received: January 20, 2016 | Revised: February 10, 2016 | Accepted: February 28, 2016 | Published Online: March 15, 2016
Author Details
( * ) denotes Corresponding author
Diabetic retinopathy (DR) is one of the pre-eminent causes of blindness among the diabetic affected patient. DR is caused by the sensitive tissue in the retina because of the damaged blood vessels. Early detection of DR can potentially reduce the risk of blindness and it could be prevented with an early screening process .In this paper authors have attempt to develop an automated exudates detection system for the classification of retinal image into exudates and non-exudates using PNN classifier. The methodologies involved in this work are Pre-processing, Optic disc elimination, Segmentation of exudates using K-means clustering technique, to classify these segmented images into exudates and non-Exudates image, a set of features based on texture and color are extracted using Gray Level Co-Occurrence Matrix (GLCM). Feature selection is done using Genetic Algorithm, and then these selected features are classified into exudates and non-exudates using Probabilistic Neural Network (PNN) classifier with an accuracy of 96.9%.
Keywords
Diabetic Retinopathy; Exudates; K-means; GLCM; Genetic Algorithm; PNN.