摘要
为了有效地提高虹膜识别系统的识别率,对Gabor滤波器进行了研究,并提出了基于层次分析法的加权Hamming距离的虹膜识别方法。首先,用不同尺度和方向的奇对称Gabor滤波器对归一化后的虹膜滤波生成特征图像;然后,对特征图像进行Hamming距离匹配得到不同滤波器的正确识别率和等错率,根据鉴别能力,选取合适的尺度和方向进行加权Hamming距离计算,其权值用层次分析法计算;最后,利用加权后的距离进行识别。其中CASIA2.0和MMU1图像库的等错率分别降低了0.64%和0.56%。实验结果表明:本文方法具有很好的识别性能和鲁棒性。
In order to improve recognition rate of the iris recognition system, the Gabor filter is studied, and the weighted Hamming distance based on analytic hierarchy process is proposed. First, the feature image is generated through different scales and directions of the odd symmetric Gabor filter. Then, the recognition correct rate and error rate are acquired by Hamming distance matching. The appropriate scale and direction are selected to calculate the weighted Hamming distance. The weight is calculated by analytic hierarchy process. Finally, the weighted Hamming distance is used to iris recognition. The equal error rate of CASIA2. 0 and MMU1 are decreased by 0. 64% and 0. 56%, respectively. The experimental results show that this method has better recognition performance and robustness.
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2014年第1期196-202,共7页
Journal of Jilin University:Engineering and Technology Edition
基金
国家自然科学基金项目(51205268)
