A Novel Retinal Blood Vessel Segmentation Method Based on Line Operator and Edge Detector

The segmented blood vessel in retinal images is an important indicator in medical treatment. In 2007, Ricci and Perfetti proposed a simple and efficient blood vessel segmentation method based on line operator. However, this scheme makes some false segmentation when it is used to process the pixels which are close to a thick blood vessel. To overcome the above problem, a novel retinal blood vessel segmentation method based on line operator and edge detector, also called (RBVSLE), was proposed. The RBVSLE enhances the contrast of the green-channel of the retinal images. Next, the RBVSLE creates the edge map, which is denoted as vessel growth seed map, by using a canny detector. Finally, the RBVSLE segments vessel pixel by line operator from vessel growth seed map. Experimental results confirm the accuracy rate of RBVSLE is approximately 94% which is significantly better than other methods.