为分割出眼底图像中的视盘, 构建基于眼底图像的计算机辅助诊断系统, 提出了一种基于视网膜主血管方向的视盘定位及提取方法。首先, 利用Otsu阈值分割眼底图像R通道获取视盘候选区域; 然后利用彩色眼底图像的HSV空间的H通道提取视网膜主血管并确定主血管方向; 在此基础上, 通过在方向图内寻找出对加权匹配滤波器响应值最高的点确定视盘中心位置; 最后, 利用该位置信息从视盘候选区域中“挑选”出真正的视盘。利用该方法对100幅不同颜色、不同亮度的眼底图像进行视盘分割, 得到准确率98%, 平均每幅图像处理时间1.3 s。结果表明: 该方法稳定可靠, 能快速、有效分割出眼底图像中的视盘。

In order to segment optic disc from fundus images, and develop a computer aided diagnosis system, an approach for location and extraction of optic disc based on direction of main blood vessel was established and studied. Firstly, red channel of the original fundus image is coarsely segmented by Otsu thresholding, and candidate regions of optic disc are obtained. Secondly, main blood vessel are extracted in the H channel of the HSV space and their directions are detected. Thirdly, the optic disc is localized by finding out the point with the top response value of the weighted matched filter. Finally, the real optic disc is extracted from candidate regions. The approach is tested on a set which contained 100 fundus images with variable color and brightness. Accuracy reaches 98%. Furthermore, the average time cost in processing an image is 1.3 seconds. Results suggest that the approach is stable and reliable, and can fast and effectively detect optic disc from fundus images.