Abstract: Detecting abnormalities such as venous looping or beading sis critical for early treatment as they are in most cases indication of potentially sight-threatening retinopathy. In order to utilize these useful characteristics of retinal blood vessels, it is very important to obtain their locations and shapes accurately. Blood vessels appeared as networks of either deep red or orange-red filaments that originated within the optic disc and were of progressively diminishing width. The major problem of Patients suffering from the disease over a long period of time is more likely to develop eye problems known as diabetic retinopathy. The detection of blood vessels is a major problem in the automatic processing of retinal images. The problem is to identify the segmentation of green veins and red veins in retina. The red and green component is extracted from the retinal images with help of KCN clustering. The accuracy detected in the green component is 96.5 % and the red component extraction is the 99.9% extractions of the retinal images.

Keywords: Veins, retina, abnormalities, Blood vessels etc.