Retinal Vessel Segmentation Based on Multidirection Filter

Abstract

In order to segment retinal vessels with complex morphology and structure, a retinal blood vessel segmentation method based on multi-direction filter is proposed. First, the image is enhanced by morphological top hat transformation and image histogram equalization. Then multi-directional Cake filtering is performed, and the filtered results are fused. Finally, the adaptive vector field divergence method is used to extract the threshold to obtain the final result. We demonstrate this using the task of segmenting blood vessels in fundus images of two standard datasets, DRIVE and STARE. The experimental results show that the proposed method adapts to the complex and changeable characteristics of retinal blood vessel scale information, and it can handle the junction of complex blood vessels with higher sensitivity.