An active contour model for segmenting and measuring retinal vessels

Al-Diri, Bashir and Hunter, Andrew and Steel, David (2009) An active contour model for segmenting and measuring retinal vessels. IEEE Transactions on Medical Imaging, 28 (9). pp. 1488-1497. ISSN 0278-0062

Documents
AlDiri2009TMIAnActiveContourModelForSegmentingAndMeasuringRetinalVessels.pdf
[img]
[Download]
Request a copy
[img] PDF
AlDiri2009TMIAnActiveContourModelForSegmentingAndMeasuringRetinalVessels.pdf - Whole Document
Restricted to Repository staff only

634kB

Full text URL: http://dx.doi.org/10.1109/TMI.2009.2017941

Abstract

This paper presents an algorithm for segmenting and measuring retinal vessels, by growing a "Ribbon of Twins" active contour model, which uses two pairs of contours to capture each vessel edge, while maintaining width consistency. The algorithm is initialized using a generalized morphological order filter to identify approximate vessels centerlines. Once the vessel segments are identified the network topology is determined using an implicit neural cost function to resolve junction configurations. The algorithm is robust, and can accurately locate vessel edges under difficult conditions, including noisy blurred edges, closely parallel vessels, light reflex phenomena, and very fine vessels. It yields precise vessel width measurements, with sub-pixel average width errors. We compare the algorithm with several benchmarks from the literature, demonstrating higher segmentation sensitivity and more accurate width measurement.

Item Type:Article
Keywords:Parametric active contour, retinal vessel segmentation
Subjects:G Mathematical and Computer Sciences > G700 Artificial Intelligence
G Mathematical and Computer Sciences > G740 Computer Vision
Divisions:College of Science > School of Computer Science
ID Code:2075
Deposited By: Bashir Al-Diri
Deposited On:02 Dec 2009 11:26
Last Modified:04 Jun 2014 15:09

Repository Staff Only: item control page