A Comparative Study of Various Exudate Segmentation Techniques for Diagnosis of Diabetic Retinopathy

Abstract

Diabetic retinopathy is a health problem which causes blindness in middle and advanced age group. Automatic identification of exudates in retinal images can contribute to early diagnosis of DR. Many approaches in literature are discussed on segmenting the exudates. This paper has shown different techniques of exudates segmentation with its benefits and limitations. All discussed techniques have improved the performance in terms of accuracy, specificity and sensitivity. The comparison has shown that ant colony optimization based segmentation has better results over each technique.