一种虹膜定位的新算法

针对虹膜外径边缘图像提取的困难,提出对增强了对比度的虹膜图像进行阈值分割之后用圆检测方法进行虹膜定位的简便而快速的算法.首先,根据虹膜图像的边缘图像用圆检测随机Hough变换方法提取瞳孔的圆心与半径;然后,对用直方图均衡化方法增强了对比度的虹膜图像进行阈值分割,提取分割后的图像的二值边缘图像;最后,利用已经提取的瞳孔的圆周参数等先验知识检测虹膜外径与圆心。实验结果表明,该算法提高了虹膜定位的速度,并且具有较好的健壮性与稳定性。

A New Iris Location Algorithm

Owing to the difficulties in extracting iris exterior radii edge image,a new simple and fast iris location algorithm based on segmenting the contrast enhanced image by threshold and circle detection is proposed.Based on this idea,circle detection random hough transform is used to extract pupil circle center and radii from the edge map of iris image;then,enhancing the original image based on histogram equalizing,segmenting the enhanced image by threshold,extracting the binary edge map of the segmented image;at last,taking advantage of the prior knowledge of the extracted pupil parameters,the iris exterior radii and circle center are detected.Experimental results show that the method can improve location speed and is not sensitive to noise effect.