基于图像抽样的快速虹膜定位算法

Rapid Iris Localization Algorithm Based on Image Sampling

提出了一种基于图像抽样的快速虹膜定位算法。首先在抽样图像中搜索瞳孔内一点,并以该点为起点检测粗略的虹膜内边缘点,然后在原分辨力图像中利用梯度算子精确定位内边缘点从而实现内边缘定位;虹膜外边缘定位采用Canny算子和Hough变换实现,由于基于抽样图像进行边缘提取,忽略了虹膜纹理等细节边缘信息,减少了大量外边缘干扰,提高了算法实时性。实验结果表明该算法的定位准确率达到99.47%,定位速度为0.162s。与经典的虹膜定位算法相比,该算法的定位速度有了很大提高。

A rapid iris localization algorithm based on image sampling was proposed. One point in the pupil was firstly detected in the sampled image. With this point as the initial one, the coarse edge points were detected, and then the inner iris edge could be located by the exact edge points extracted by gradient operator in the image with original resolution. The outer iris edge was located using Canny operator and Hough transform. Since the edges were extracted based on the sampled image, plenty of detailed boundaries such as iris texture were ignored, which reduced the outer disturbing boundaries and improved the real-time property of algorithm. The experiment results show that the algorithm achieves 99.47 % in accuracy and 0.162 s in speed. In comparison with other classical localization methods, the developed algorithm is much faster.