Multi-scale approach for retinal vessel segmentation using medialness function

Automated segmentation of retinal vessels in optic fundus images has been the most prevailing effort in many researches during recent years. In this paper, we propose a multi-scale method based on a weighted 2D medialness function. The result of the medialness function is first multiplied by the eigenvalues of the Hessian matrix in every pixel of the image in order to extract vessel's medial-lines. Next, by extracting the centerlines of vessels and estimation of radius of vessels, the retinal vessels are segmented. Finally, the performance of our proposed method is evaluated by the DRIVE and STARE databases and compared with those of several recent methods.