Blood Vessel Segmentation Of Retinal Images Using Cricket Chirping Algorithm

Abstract

Blood vessels in the retina play a crucial role in diagnosing and treating diabetic retinopathy, Aneurysms, and cardiovascular diseases. Regular examination of retinal images in the patient can facilitate early identification of diabetic retinopathy and other ophthalmological diseases, thereby preventing the occurrence of blindness and other severe complications. Injuries or medical conditions like hypertension, diabetes, etc., can change the structure of blood vessels near to the retina.  Generally, the retinal image is  composed of  thin end vessels  and hence difficult to identify. Manual segmentation of retinal layers is tedious, time consuming, and subjective. Hence several retinal segmentation algorithms were employed for rapid and accurate delineation of blood vessels from several retinal layers assisting in identifying retinal disease. The image segmentation process helps to segment the blood vessels in the fundus image and provides easy processing. Several nature-inspired algorithms were used for image segmentation. In this paper cricket chirping algorithm (CCA) considering Otsu’s between class-variance method as objective function was used for retinal image segmentation. This method showed promising results when tested on the DRIVE and STARE datasets. This method can act as a diagnostic decision-making system for detecting retinal diseases.