基于相位一致性的眼底图像视网膜血管分割  

Retinal vasculature's segmentation in fundus image based on phase congruency

作　　者：陶治江[1] 高原[2] 黄华[1] 

机构地区：[1]四川大学电气信息学院,四川成都610065 [2]中国医学科学院北京协和医院,北京100730

出　　处：《计算机工程与设计》2012年第8期3139-3143,共5页Computer Engineering and Design

摘　　要：眼底图像的视网膜血管分割是眼底图像处理的重要组成部分,视网膜血管对于医学研究和临床诊断有着重要的作用。传统图像分割算法都有一定的缺陷,而相位一致性算法由于不受对亮度和对比度的影响,且有着较好的分割效果,可以用于图像特征的提取和分割。为此提出了将相位一致性算法应用于眼底图像的血管提取中,采用真实的眼底图像数据库进行实践,证明了可较好地用于眼底图像视网膜血管分割。Retinal vasculature＇s segmentation in fundus image takes one of the most important roles in fundus image process, and contributes greatly to the medical research, clinical diagnosis and treatment of many diseases. Some traditional image segmentation algorithms are good, but also have their own defects, while phase consistency algorithm takes the advantage of excellent segmentation effect without the affect of illumination and contrast, which is easily used in image feature extraction and segmentation. The segmentation of the retinal vessels by the phase congruency algorithm is proposed and practiced with the real fundus image database. It prove the phase congruency is well used in the retinal vasculature＇s segmentation.

关 键 词：相位一致性 图像分割 眼底图像 视网膜血管 logGabor小波 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]