基于灰度-梯度共生矩阵的视网膜血管分割方法

提出了一种新的、有效的视网膜血管分割方法．它包括二维匹配滤波预处理以增强血管的灰度，以及用灰度-梯度共生矩阵的最大熵阈值化方法．该方法同时利用了图像的灰度和梯度信息．在计算梯度时选用了三次B样条小波．实验结果表明，该方法能较好地提取视网膜血管网络．

The Segmentation Method of Retinal Blood Vessels Based on Gray Level-Gradient Co-occurrence Matrix

This paper presented an efficient method for segmentation blood vessels in retinal image. The proposed algorithm is composed of two-step: two-dimension matched filtering, entropy thresholding. The purpose of the matched filtering is to enhance the blood vessels. Entropy thresholding method is an automatic technique for thresholding of digital images based on gray level-gradient co-occurrence matrix and the maximum entropy principle. This method attempts to utilize the information of both gray level and gradient in an image. Cubic B-spline wavelet is used to calculate the gradient value of this co-occurrence matrix. The algorithm was tested on some ocular fundus images, and better experimental results were showed.