Paper Info
Title
Method for retinal vessel detection and diameter measurement
Abstract
Since retinal vessel detection and measurement plays an important role in diagnosing cardiovascular diseases, a lot of efforts have been made in recent years. In this paper, we propose an efficient method for vessel detection and diameter measurement which incorporates with edge detection, path searching, and matched filtering. Firstly, Gaussian-Laplacian filter and zero crossing detection are applied to retinal image to get an initial edge-map. Then, for any vessel of interest, its centerline and diameter can be accurately obtained by using region growing, thinning, path searching, and matched filter. The robustness and efficiency of the proposed method have been demonstrated by our extensive experiments. Another advantage is that it can process retinal image in real-time speed. The detection and measurement results lead us to determine more parameters for further study of vascular diseases.
Year
DOI
Venue
2004
10.1117/12.534539
Proceedings of SPIE
Keywords
Field
DocType
computer vision,medical imaging,image processing and analysis,computer-aided diagnosis
Computer vision,Zero crossing,Computer science,Edge detection,Computer-aided diagnosis,Filter (signal processing),Robustness (computer science),Gaussian,Region growing,Artificial intelligence,Matched filter
Conference
Volume
ISSN
Citations 
5370
0277-786X
0
PageRank 
References 
Authors
0.34
0
4
Name
Order
Citations
PageRank
xiaokun li141.09
Tong Fang2144.44
Jenn-Kwei Tyan3566.68
ming fang4133.84