An automated method using hessian matrix and random walks for retinal blood vessel segmentation

Abstract

Retinal blood vessel detection is very significant in the disease diagnosis made by ophthalmologists. Thus, an automated method to segment the blood vessels in retinal images is proposed using the hessian-based filter and random walk algorithms. The aim of the hessian-based vascular filtering is to enhance the vessel structures. Local thresholding is also used to obtain seed groups for the random walk segmentation. Experiments on the DRIVE (Digital Retinal Images for Vessel Extraction) database clearly show that the method proposed in this paper can achieve better performance. The results are compared to other retinal blood vessel segmentation methods with respect to the accuracy, sensitivity and specificity.