Abstract: In this paper an algorithm for vessel segmentation in retinal images is proposed. Firstly, on the basis of preprocessing of the retinal image, the contrast between retinal image vessel and background is improved. Then the multi-scale enhancement filter based on the Hessian matrix is used to enhance the retinal image, and finally the whole vessel network is binarized by using an iterative thresholding method.
The experimental evaluation in the publicly available DRIVE database shows accurate extraction of vessels network. The average accuracy of 95.88 with 0.729 true positive rate and 0.194 false positive rate, which is very close to the manual segmentation rates obtained by the second observer. The proposed algorithm is compared also with widely used supervised and unsupervised methods, and the experimental results show the effectiveness of the method.
Keywords: Retinal image; Preprocessing; Hessian matrix; Iterative thresholding.
Title: Multi-Scale Retinal Vessel Segmentation Using Hessian Matrix Enhancement
Author: Ning Cui, Xiaoting Liu, Song Yang
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
