Abstract
Fundus examination is a non-invasive procedure of observing changes in retinal vasculature linked with the identification and progression of certain ocular diseases. Segmenting vessels from the rest of the structure is found helpful in analyzing and later tracking the changes. Manual vessel segmentation requires clinical expertise, and with large scale screening certainly puts a burden on already scarce clinical resources. A computer-aided diagnosis (CAD) recently emerged to alleviate this burden. A variety of computerized methods have emerged recently with the primary aim of providing accurate vessel segmentation. One particularly interesting approach is multi-scale line filtering. However, its response diminishes in low-contrast areas of the image causing certain vessels to be missed. In this paper, we investigate the use of phase symmetry detector to get help with low-contrast vessel detection. This specific detector does not make any assumptions about the luminance profile of the vessel but then has major drawback of being sensitive to background noise. To reduce the noise sensitivity, we adopted the multi-scale line filtering with an improved vessel uniformity function as an input to the phase symmetry detector. The low-contrast vessel information thus made available helps in providing an improved accuracy for automated vessel segmentation algorithms. The quantitative tests are conducted for two publicly available databases (DRIVE, STARE) of fundus images that shows promise of improvements in all three performance categories called accuracy, sensitivity, and specificity.