基于自适应多尺度模板匹配的视盘检测方法

针对视盘检测易受光照和弱对比度影响的问题,提出了一种全新的视盘检测方法用于有效地定位和分割视盘.首先,采用预处理技术校正不均匀的光照和提高弱的对比度.然后,利用交替序列滤波和区域极大值技术提取一系列的视盘关键点.再次,利用提出的自适应多尺度模板匹配方法,计算每一个视盘关键点的相关系数,并将最大相关系数值所对应的关键点视为视盘中心.最后,基于获得的视盘中心,提取包含该中心位置的感兴趣区域,并在此基础上,利用Canny边缘检测算子和霍夫变换技术,实现视盘边缘的有效估计.该算法在DRIVE、DIRATEDB0、DIRATEDB1和ROC四个公共数据库上进行了测试,实验结果表明,提出算法的性能明显地优于现有方法.

Optic Disk Detection Approach Based on Adaptive Multi-scale Template Matching

To solve the problem of optic disk detection being easily affected by light and weak contrast,a novel approach for automatic optic disk detection is proposed for effectively locating and segmenting the optic disk.First,preprocessing is applied to correct the uneven illumination and improve the low contrast.Then,a series of key points can be extracted using alternative sequential filters and regional maxima.Moreover,the key point that has the maximum correlation coefficient calculated by adaptive multi-scale template matching method is located as the center of the optic disc.Finally,the region of interest that contains the optic disc location is extracted,and the optic disc boundary can be effectively estimated using Canny edge detection and Hough transform technique.The proposed approach is tested on four publicly available databases,namely,DRIVE,DIRATEDB0,DIRATEDB1 and ROC.Experimental results show that the proposed approach performs better than state-of-the-art approaches.