Paper Info
Title
Retinal Vessel Segmentation Using the 2-D Morlet Wavelet and Supervised Classification
Abstract
We present a method for automated segmentation of the vasculature in retinal images. The method produces segmentations by classifying each image pixel as vessel or non- vessel, based on the pixel's feature vector. Feature vectors are composed of the pixel's intensity and continuous two-dimen sional Morlet wavelet transform responses taken at multiple scales. The Morlet wavelet is capable of tuning to specific frequencies, thus allowing noise filtering and vessel enhancement in a single s tep. We use a Bayesian classifier with class-conditional probabi lity density functions (likelihoods) described as Gaussian mixtures, yielding a fast classification, while being able to model com plex decision surfaces and compare its performance with the linear minimum squared error classifier. The probability distribu tions are estimated based on a training set of labeled pixels ob- tained from manual segmentations. The method's performance is evaluated on publicly available DRIVE (1) and STARE (2) databases of manually labeled non-mydriatic images. On the DRIVE database, it achieves an area under the receiver operating characteristic (ROC) curve of 0.9598, being slightly superior than that presented by the method of Staal et al. (1).
Year
Venue
Keywords
2005
Clinical Orthopaedics and Related Research
morlet,index terms— fundus,pattern classification,vessel segmentation,wavelet.,retina,roc curve,receiver operator characteristic,indexing terms,bayesian classifier,feature vector,wavelet transform
DocType
Volume
Citations 
Journal
abs/cs/051
10
PageRank 
References 
Authors
0.99
21
5
Name
Order
Citations
PageRank
João V. B. Soares155121.31
Jorge J. G. Leandro239917.92
Roberto M. Cesar, Jr.379449.46
Herbert F. Jelinek447736.78
Michael J. Cree561745.67