计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (17): 185-187.

• 图形、图像、模式识别 • 上一篇    下一篇

基于改进的有监督保局投影人脸识别算法

高 雷,李晓东   

  1. 临沂大学 信息学院,山东 临沂 276005
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-06-11 发布日期:2011-06-11

Improved LPP algorithm for face recognition

GAO Lei,LI Xiaodong   

  1. School of Information,Linyi University,Linyi,Shandong 276005,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-06-11 Published:2011-06-11

摘要: 为了充分利用样本的类别信息,提出了一种改进的有监督保局投影人脸识别算法。利用先验类标签信息重新构造传统保局投影算法中的权重矩阵,基于改进后的保局投影算法得到变换矩阵;用线性鉴别的思想筛选出变换矩阵中的最优基向量,构成最终的变换矩阵。把训练样本和测试样本投影到由最优基向量构成的子空间得到训练样本和测试样本的特征。采用最近邻分类器分类。在ORL和FERET人脸库上的测试结果表明,算法具有较好的识别性能。

关键词: 人脸识别, 保局投影, 线性鉴别

Abstract: In order to make full use of the classification information of samples to get optimal features,a new supervised LPP algorithm for face recognition is proposed.Between-class scatter matrix is embedded in the objective function of original locality preserving projections,and the transformation matrix can be obtained based on the modified objective function.Subsequently,according to the idea of linear discriminant,the optimal base vectors of the transformation matrix are selected to form the final transformation matrix.As a result,the features of training samples and testing sample are got by projecting them on the subspace spanned by optimal base vectors.Finally,Nearest Neighborhood(NN) algorithm is used to construct classifiers.Experiment on ORL and FERET face databases is conducted to demonstrate recognition performance of the method,and the results show that it is effective.

Key words: face recognition, locality preserving projections, linear discriminant