One of the most commonmodalities to examine the human eye is the eye-fundus photograph.Theevaluation of fundus photographs\r\nis carried out by medical experts during time-consuming visual inspection. Our aim is to accelerate this process using computer\r\naided diagnosis. As a first step, it is necessary to segment structures in the images for tissue differentiation. As the eye is the only\r\norgan, where the vasculature can be imaged in an in vivo and noninterventional way without using expensive scanners, the vessel\r\ntree is one of the most interesting and important structures to analyze. The quality and resolution of fundus images are rapidly\r\nincreasing.Thus, segmentation methods need to be adapted to the new challenges of high resolutions. In this paper, we present a\r\nmethod to reduce calculation time, achieve high accuracy, and increase sensitivity compared to the original Frangi method. This\r\nmethod contains approaches to avoid potential problems like specular reflexes of thick vessels. The proposed method is evaluated\r\nusing the STARE and DRIVE databases and we propose a new high resolution fundus database to compare it to the state-of-theart\r\nalgorithms.The results show an average accuracy above 94% and low computational needs.This outperforms state-of-the-art\r\nmethods.
Loading....