Retinal Blood Vessel Segmentation Using Line Operators and Support Vector Classification

Abstract

In the framework of computer-aided diagnosis of eye diseases, retinal vessel segmentation based on line operators is proposed. A line detector, previously used in mammography, is applied to the green channel of the retinal image. It is based on the evaluation of the average grey level along lines of fixed length passing through the target pixel at different orientations. Two segmentation methods are considered. The first uses the basic line detector whose response is thresholded to obtain unsupervised pixel classification. As a further development, we employ two orthogonal line detectors along with the grey level of the target pixel to construct a feature vector for supervised classification using a support vector machine. The effectiveness of both methods is demonstrated through receiver operating characteristic analysis on two publicly available databases of color fundus images.

Retinal Blood Vessel Segmentation Using Gabor Wavelet and Line Operator

“…As the result shows the combination of Gabor features and line features improves the performance. We observed the result of image classification using [1] and [2] to find the reason for method.…”

“…In [2], those pixels located between two thick vessels are misclassified as vessel while they belong to non vessel class (Fig. 2).…”

“…The basic line operator is a line with length l [2] centered at considered pixel. At each pixel the average of image gray level along line operators with 12 different orientations spanning 360 degrees are evaluated.…”

“…In [1] a Bayesian classifier in [10], [11] and [12] a K-Nearest Neighbor and in [2] Support Vector Machine have been used as classifiers for pixel segmentation. In classification process each pixel of retinal image takes one of two labels: vessel or non-vessel.…”

“…These two characteristics are specially suited for the detection of directional features and have been chosen in order to enable the transform to present stronger responses for pixels associated with the blood vessels. Basic line detector has used with 15 = l pixels length [2]. In the first experiment 700,000 pixel samples were randomly chosen from 20 labeled training images to estimate density function of vessel and non vessel using GMM.…”