视网膜血管分割在眼底图像分析中具有重要作用。结合多尺度Hessian矩阵滤波和线检测算子, 提出了一种有效的血管检测方法。首先利用多尺度Hessian矩阵的特征值构建血管相似性函数, 实现血管增强; 然后采用改进的线检测算子, 提取反映血管测度的特征;最后采用SVM实现血管检测。实验结果表明, 该方法只需要较少的训练样本即可达到与其他方法相当的准确率, 且在灵敏性上具有更好的性能。

Retinal blood vessel segmentation plays an important role in the analysis of retinal image. An effective method for vessel segmentation is proposed based on multi-scale Hessian matrix filtering and line detector. First, the eigenvalues of the Hessian matrix are used to build blood vessels’similarity function to enhance blood vessels. Then, the improved line operator is adopted to extract the vascular feature .At last, vascular detection is realized by using SVM. The results show that the method can achieve considerable accuracy and better performance on sensitivity compared with other methods, only with a small number of training samples.