基于小波包分解的虹膜识别算法  

Iris recognition algorithm based on wavelet packet decomposition

作　　者：孟爱国[1] 章登勇[1] 李峰[1] 

机构地区：[1]长沙理工大学计算机与通信工程学院,湖南长沙410076

出　　处：《计算机工程与设计》2005年第10期2756-2758,共3页Computer Engineering and Design

摘　　要：特征提取是虹膜识别的关键技术;由于虹膜图像具有丰富的纹理,提出了基于小波包分解的虹膜识别算法。小波包分解不仅包含了图像的低频部分而且还保留了高频部分,它能够有效地提取虹膜纹理特征,并按hamming距离对虹膜进行匹配。实验结果表明,该算法具有良好的鲁棒性。Feature extraction is regarded as a key technology of iris recognition; owing to ample texture existing in iris image, an iris recognition algorithm based on wavelet packet decomposition is proposed. Wavelet packet decomposition retains not only the low but also the high frequency part of iris image, it can effectively extract texture features of iris, and matches the iris by means of hamming distance. The results of the experiment show that this algorithm has good robustness.

关 键 词：虹膜识别 特征提取 小渡包分解 

分 类 号：TP391.14[自动化与计算机技术—计算机应用技术]