Retinal blood vessel segmentation by support vector machine classification

Medical diagnostics has been significantly improved by introduction of digital imagery, primarily because of the powerful digital image processing tools. Digital retinal images are used for diagnostics of various diseases including diabetes, hypertension, stroke, etc. Retinal blood vessels are crucial for such diagnostics so segmentation of retinal blood vessels is an important and active research area. In this paper we propose an overlapping-block-based algorithm for retinal blood vessels segmentation based on classification by support vector machine using chromaticity and DCT coefficients as features. The proposed algorithm was tested on standard benchmark retinal images from the DRIVE data set. Results were compared with available ground truth images and other approaches from literature and vessel segmentation was excellent in all cases.