Algorithm for Fundus Image Analysis through Segmentation and Object Classification of Tracing Retina Blood Vessel

Abstract

Fundus imaging is complicated by the fact that the illumination and imaging beams cannot overlap because that results in corneal and lenticular reflections diminishing or eliminating image contrast. Consequently, separate paths are used in the pupillary plane, resulting in optical apertures on the order of only a few millimeters. Fundus imaging is the most established way of retinal imaging. Until recently, fundus image analysis was the only source of quantitative indices reflecting retinal morphology. The major limitation of fundus photography is that it obtains a 2-D representation of the 3-D semi-transparent retinal tissues projected onto the imaging plane. The initial approach to depict the 3-D shape of the retina was stereo fundus photography The proposed algorithm was verified by using two online databases, DRIVE and HRF to validate the performance measures. Hence, proposed method is capable to extract the retina blood vessel and give the accuracy of 0.7917, the sensitivity of 0.9077 and the specificity of 0.7215. In conclusion, the extraction of the blood vessels is highly recommended as the early screening stage for the eye diseases beneficially.