基于巴氏距离和LPP相结合的人脸识别  

Face recognition based on Bhattacharyya distance and locality preserving projection

作　　者：任成娟[1] 

机构地区：[1]宝鸡文理学院计算机科学系,陕西宝鸡721000

出　　处：《电子设计工程》2013年第11期16-18,共3页Electronic Design Engineering

基　　金：宝鸡文理学院院级重点项目(ZK10168)

摘　　要：局部保持映射(Locality Preserving Projection,LPP)算法是一种有效的特征提取方法。提出了利用巴氏距离和LPP相合算法对特征进行提取。当特征维数过高时,首先对样本用LPP进行特征提取和降维处理,然后采用巴氏距离特征的迭代算法,得到最小错误率上界。在ORL上实验,实验结果表明了提出算法在人脸识别中的有效性。Locality Preserving Projection is an effective method which can extract the feature. This paper proposes a smart feature selection method which combines principal component analysis and Bhattacharyya distance. When the feature vector dimension is high, firstly the sample dimension is reduced by using LPP. Then it gets the smallest error rate upper bound by using iterative algorithm which has Bhattacharyya distance feature. The experimental results demonstrate the efficacy of the proposed approach for face recognition on ORL.

关 键 词：局部保持映射 巴氏距离 特征提取 人脸识别 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]