中文摘要：

凭借虹膜A身的稳定性、非侵犯性、难更改性等优点，虹膜识别已经成为生物特征识别领域中的研究热点。但是在非侵犯性的虹膜识别系统所采集到的人眼图像中，虹膜环内的部分区域被眼睑或眼毛所遮挡，造成虹膜纹理的丢失。将被眼睑或眼毛遮挡的区域去掉之后，虹膜可用纹理区域的面积会低于100％。为了在这种情况下仍然能够提取出满足虹膜识别要求的特征，本文提出了一种基于相位一致性的虹膜识别方法。实验结果证明，该方法能够在虹膜可用纹理区域不完整的情况下提取出足够多的可区分性特征用于虹膜识别，并且具有0．1370％的等错率和99．8625％的正确识别率。

英文摘要：

Thanks to many advantages of the iris, such as stable, nonintrusive and difficult to be changed, iris recognition has been a research hot-spot in biometrics personal identification field. However, in the eye image captured by nonintrusive iris recognition system, some regions of the iris annulus are covered by eyelids or eyelashes and many veins are destroyed. When these regions covered by eyelids or eyelashes are wiped off, the area of usable iris region will be less than 100%. In order to extract sufficient features to satisfy the requirement of iris recognition, a method based on phase congruency is proposed. Experimental results demonstrate that the proposed method can extract e- nough distinguishable features in the incomplete usable iris region for iris recognition, and the method has the equal error rate of 0. 137 0% and the correct recognition rate of 99. 862 5%.