一种新的基于小波变换的虹膜识别算法  

Wavelet transform algorithm for iris recognition

作　　者：黄惠芳[1] 胡广书[1] 

机构地区：[1]清华大学生物医学工程系,北京100084

出　　处：《清华大学学报（自然科学版）》2003年第9期1210-1213,共4页Journal of Tsinghua University(Science and Technology)

摘　　要：为了克服基于小波变换过零检测的虹膜识别算法对纹理灰度变化敏感的缺点,利用小波的带通特性和小波变换检测奇异点的原理,提出了一种新的基于小波变换的虹膜识别算法。将虹膜纹理分成8个分析带,每个分析带只采用一个合适尺度的小波滤波,然后将滤波结果量化,产生256byte的虹膜特征码,最后利用Hamming距离进行模式匹配。实验结果表明,算法运算时间约为1s,识别率可达97.04%,可用于实际的身份鉴别系统中。同时,算法具有平移、旋转和缩放不变性。Iris recognition algorithms based on wavelet transform zerocrossing are sensitive to the grey variation of the iris image texture. A new iris recognition algorithm was developed using the bandpass characteristic of wavelets and the wavelet transform principles for detecting singularities. The iris image texture is divided into eight bands with an appropriate scale wavelet used to filter each band for quantization to 256 byte iris feature codes. Then the Hamming distance is used to match two iris codes. Tests show that this method which is translation, rotation, and scaling invariant requires approximately 1s to complete recognition and achieves an identification rate of 97.04%, so it can be used in a personal identification system.

关 键 词：虹膜识别 小波变换 特征提取 HAMMING距离 模式匹配 身份鉴别系统 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]