Segmentation of Retinal Blood Vessels Based on Transition Region Extraction
Segmentation of Retinal Blood Vessels Based on Transition Region Extraction
复制标题
DOI：
--
复制
发表时间：
2008
期刊：
影响因子：
--
通讯作者：
Liu Ju-peng
中科院分区：
文献类型：
--
作者：
Liu Ju-peng
针对现有视网膜血管分割方法对小血管和低对比度血管分割效果不佳的问题，提出了一种基于过渡区提取的视网膜血管分割方法。该方法首先利用二维高斯匹配滤波器对视网膜图像进行增强，然后利用最优熵法分割出主要血管，并利用基于分布式遗传算法和大津算法的过渡区提取算法，最后，对过渡区进行分割。在Hoover数据库上进行的实验表明，该方法在小血管提取、连通性和有效性方面均优于Hoover算法，并且引入了基于迁移策略的分布式遗传算法，提高了该方法的效率.
Aiming at the bad performance of existing retinal blood vessel segmentation methods for small and low contrast vessels,a new segmentation method based on transition region extraction is proposed.Firstly,the two-dimensional Gaussian matched filter is used to enhance the retinal image.Then the main vessels are segmented through optimal entropy method and the transition region is extracted by algorithm based on distributed genetic algorithm and Otsu.Finally,the vessels are obtained via analyzing the region connectivity.The experiments implemented on the Hoover database indicate that the proposed method outperforms the Hoover algorithm on the small vessels extraction,connectivity and effectiveness.In addition,the efficiency of this method could be improved by introducing the distributed genetic algorithm based on migration strategy.