计算机应用与软件2012，Vol.29Issue(3)：270-273,293,5.
基于分类性能的掌纹特征有机融合
PALMPRINT FEATURES FUSION IN ORGANIC WAY BASED ON CLASSIFICATION PERFORMANCE
李云峰 1张亚莉1
作者信息
- 1. 河南科技大学机电工程学院,河南洛阳471003
- 折叠
摘要
Abstract
In order to overcome the limitation of palmprint recognition by using single feature, a method of palmprint multi-feature fusion in an organic way based on classification performance is proposed in this paper. This features fusion method is realised in three steps: the feature selection, the weighting processing and the dimension reduction processing. Based on analysing the clustering performance of each single fea ture component, the features' trade-off is carried out by constructing the criterion and set the corresponding threshold value; the weight value of each feature component is calculated by constructing the function of within-class distance and between-class distance, and the weighting processing is then performed; at last the principal components analysis (PCA) is used to reduce the dimension of the feature. The improved Local Binary Patterns (LBP) and Discrete Wavelet Transform ( DWT) are used to extract the two kinds of palmprint features, and these ex tracted features are used in fusion experiments, the results show the effectiveness of the method.关键词
掌纹识别/特征融合/分类性能/LBP/DWTKey words
Palmprint recognition/Feature fusion/Classification performance/LBP/DWT分类
信息技术与安全科学引用本文复制引用
李云峰,张亚莉..基于分类性能的掌纹特征有机融合[J].计算机应用与软件,2012,29(3):270-273,293,5.
