Detection and diagnosis of blood vessel in retinal diseases

Retinal images is an methodology for diabetic retinopathy detection and analysis. The detection of blood vessels from the retinal images is a exhausting process. In this work a new algorithm to detect the blood vessels effectively has been proposed. Proposed system consists of three stages-first is preprocessing of retinal image to separate the green channel, second stage thresholding, third stage image masking, forth stage morphological processing. Initially enhancement of the image is using thresholding and changes of the variation. Since the blood vessels are distributed in different directions, morphology operate with multidirectional structuring elements are used to extract the blood vessel from the retinal images. Morphological operator by reconstruction using structure elements eliminates the ridges not belonging to the vessel. Finally applying median filter on the connected components all of images.