Retinal vessel enhancement and extraction based on directional field

Abstract

In order to extract the low contrast vessels in retinal images, we propose a novel enhancement algorithm based on directional field and develop a robust retinal vessel extraction framework. The proposed enhancement algorithm can significantly improve the contrast of vasculature especially the slim vessels and low contrast vessels. In our framework, we first enhance the vasculature by the proposed enhancement algorithm, and extract the vessels from the enhanced image by a local extraction method. Secondly, we apply a series of modified top-hat transform to original image, and use a binary morphological reconstruction to extract the large size vessels. Finally, the whole vasculature is provided by an organic combination algorithm. The first step extracts the slim vessels and low contrast vessels well but misses some bifurcations and cross pixels. On the other hand, the second step preserves the bifurcations and cross pixels but misses some slim vessels and low contrast vessels. Therefore, the combination of the two methods can provide satisfactory extraction of vasculature. Experimental results show that our method outperforms existing algorithms in terms of sensitivity and accuracy when testing on 'DRIVE' and 'STARE' databases.