Detection of Vascular Bifurcation in Retina Fundus Image for Person Identification

A method is presented for automated segmentation of vessels in two-dimensional color images of the retina. This method can be used in computer analyses of retinal images, e.g., in automated screening for diabetic retinopathy. Here we are trying to detect of vascular bifurcation from a fundus image. Vascular bifurcation can be used as one of the key characteristics to detect individual retina which can be used for an authentication system as like finger print. The acquired images undergo preprocessing stage truncation thresholding, edge detection and noise removal and skeletonization. This system used a 5x5 window probe which traverses within the image considering every pixel in the image, collecting its 16 neighboring pixels and stores the value in an array. Then the algorithm counts the black region for bifurcation or cross over point detection. By using these methods, we can get rid from the mazy angle between the retinas blood vessels. This system was tested on a database of 20 fundus image. Result obtained from applying this method after doing some necessary modification on the fundus image gives almost accurate results for every time.