| 基于小波包分析的虹膜特征提取方法 |
| |
| 引用本文： | 王琪,费耀平. 基于小波包分析的虹膜特征提取方法[J]. 计算机工程与应用, 2006, 42(13): 60-62 |
| |
| 作者姓名： | 王琪 费耀平 |
| |
| 作者单位： | 中南大学信息科学与工程学院,长沙,410075;中南大学信息科学与工程学院,长沙,410075 |
| |
| 摘 要： | 提出了一种新的虹膜特征提取方法。针对虹膜纹理图像的特征表示主要集中在中高频能量部分,对图像进行二级小波包分解,并提取出第一级和第二级的中高频部分的能量分量作为特征向量,并采用欧几里德距离作为模式识别方法。克服了用小波分解只提取虹膜的低频部分,从而不能充分地反映纹理的特征的缺陷。试验结果和数据表明了此算法的合理性和有效性。
|
| 关 键 词： | 虹膜识别 特征提取 纹理分析 小波包分解 欧几里德距离 |
| 文章编号： | 1002-8331-（2006）13-0060-03 |
| 收稿时间： | 2005-07-01 |
| 修稿时间： | 2005-07-01 |
Method for Feature Extraction of Iris Based on Wavelet Package Analysis |
| |
| Wang Qi,Fei Yaoping. Method for Feature Extraction of Iris Based on Wavelet Package Analysis[J]. Computer Engineering and Applications, 2006, 42(13): 60-62 |
| |
| Authors: | Wang Qi Fei Yaoping |
| |
| Affiliation: | School of Information Science and Engineering, Central South University, Changsha 410075 |
| |
| Abstract: | A new method for feature extraction of iris is proposed.Arming at the feature of texture is focused on middle and high frequency components,the method makes two-dimension wavelet package decomposition towards the image,extracts the middle and high frequency components of the first and second rank as eigenvector and uses Euclidean distance as the way of model identification.The method avoids the disadvantage that using the wavelet decomposition can only extract the low frequency components,which cannot reflect the feature of texture images.The experiment results and data analysis are given to indicate the effectiveness and the rationality of the algorithm. |
| |
| Keywords: | iris identification feature extraction texture analysis wavelet package decomposition euclidean distance |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
