Retinal Vessel Skeletonization Using Scale-space Theory

In this paper we introduce the alternative method for blood vessel extraction based on scale space theory. The original image is converted into gray level image and it is then blurred with Gaussian Blur using many kernel sizes. Each kernel produces an image at that particular scale. Edge detection is applied to each result using Laplace algorithm. Noises are then removed using adaptive median filter. Images are converted to binary images and the isolated islands are removed using region glowing technique. Candidate vessels from all scales are combined for the final result. The algorithm was tested on 100 images and the results are compared with ophthalmologists’ hand-drawn ground truth. The performance is very encouraging and it can detect the blood vessel with a high specificity of 0.9698.