视网膜血管图像特征点自动提取和分类

Automatic detection and classification of feature points from retinal fundus images

由动静脉血管组成的眼底视网膜血管结构的特征点是预测心血管疾病、图像分析和生物学应用的重要特征。把角点检测引入到视网膜血管分叉点和交叉点提取中,利用边缘检测算子得到二值边缘图像,采用基于累加点到弦的距离(CPDA)的角点检测方法得到候选特征点,再根据视网膜血管图像的拓扑结构设计自适应矩形探测器对候选特征点进行删减和分类。实验结果表明,基于CPDA的角点检测和自适应矩形探测器的方法有效地实现了节点的提取和分类。

The feature points of the arteriovenous retinal vessel are important landmarks in predicting cardiovascular disease,image analysis and biometrics application.Corner detection method is introduced to extract the vascular bifurcations and crossovers points from eye fundus images.The edge detection operator is used to get the binary edge image.A corner detection method based on Chord-to-Point Distance Accumulation（CPDA） technique is applied to obtain the candidate feature points.According to the topological structure of vessel,the adaptive rectangular detector is designed to classify candidate points.The results show that the method based on CPDA and adaptive rectangular detector is feasible to detect and classify feature points.