4 February 2011 An efficient algorithm for measurement of retinal vessel diameter from fundus images based on directional filtering
Author Affiliations +
Proceedings Volume 7752, PIAGENG 2010: Photonics and Imaging for Agricultural Engineering; 77521N (2011) https://doi.org/10.1117/12.888298
Event: International Conference on Photonics and Image in Agricultural Engineering (PIAGENG 2010), 2010, Qingdao, China
Abstract
Automatic measurement of vessels from fundus images is a crucial step for assessing vessel anomalies in ophthalmological community, where the change in retinal vessel diameters is believed to be indicative of the risk level of diabetic retinopathy. In this paper, a new retinal vessel diameter measurement method by combining vessel orientation estimation and filter response is proposed. Its interesting characteristics include: (1) different from the methods that only fit the vessel profiles, the proposed method extracts more stable and accurate vessel diameter by casting this problem as a maximal response problem of a variation of Gabor filter; (2) the proposed method can directly and efficiently estimate the vessel's orientation, which is usually captured by time-consuming multi-orientation fitting techniques in many existing methods. Experimental results shows that the proposed method both retains the computational simplicity and achieves stable and accurate estimation results.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xuchu Wang, Xuchu Wang, Yanmin Niu, Yanmin Niu, } "An efficient algorithm for measurement of retinal vessel diameter from fundus images based on directional filtering", Proc. SPIE 7752, PIAGENG 2010: Photonics and Imaging for Agricultural Engineering, 77521N (4 February 2011); doi: 10.1117/12.888298; https://doi.org/10.1117/12.888298
PROCEEDINGS
4 PAGES


SHARE
Back to Top