Comparative study of retinal vessel segmentation methods

Diabetic Retinopathy is one of the important complications of the diabetes mellitus, which causes serious damage to the retina leading to visual damage. Retinal vessel segmentation plays an important role in automatic retinal disease screening systems. An accurate segmentation is mandatory for the pathological analysis. To support this, several automated systems were designed in the past years. This paper analyses and compares five different segmentation methods, which exist in the literature. The performance of the algorithms is tested with the images available in DRIVE and STARE databases.