利用符号假设检验比较两类PCA人脸识别算法  

Comparison between Two Kinds of PCA-Based Face Recognition Algorithms by Sign Hypothesis Testing Strategy

作　　者：李乐[1] 章毓晋[2] 

机构地区：[1]清华信息科学与技术国家实验室,北京100084 [2]清华大学电子工程系,北京100084

出　　处：《模式识别与人工智能》2008年第4期432-440,共9页Pattern Recognition and Artificial Intelligence

基　　金：国家自然科学基金(No.60573148);教育部高等学校博士学科点专项科研基金(No.SRFDP-20060003102)资助项目

摘　　要：基于 PCA 人脸识别算法分为基于经验 PCA(EPCA)算法和基于自适应 PCA(APCA)算法.本文分析它们各自的构造原则和应用特点,借助3个公共人像数据库,用1个新构造的符号假设检验策略对它们进行客观的实验比较.比较结果显示,就整体性能而言,如果做 EPCA 训练的图像与样例图同身份,基于 EPCA 算法和基于 APCA算法间差异很小,否则,差异很大.就可得的最优性能而言,两类算法间无显著差异.基于上述结论,文中分析和解答一些较有实际意义的问题,这为深入理解和合理使用基于 PCA 人脸识别算法提供有益参考.PCA-based face recognition algorithms are actually classified into adaptive PCA based （APCA- based） algorithms and empirical PCA-based （EPCA-based） algorithms. The design principles and application characteristics of these two kinds of algorithms are analyzed. A new sign hypothesis testing strategy is designed to make objective comparisons between them on three common face databases. Two basic conclusions are drawn according to the comparison results. On one hand, as far as holistic performance is concerned, the difference between EPCA-based algorithms and APCA-based algorithms is relatively small if the training images have the same identity set as the gallery ones. Otherwise, the difference between them is very large. On the other hand, as far as the best realizable performance is concerned, there is no significant difference between them. Thus, some practical problems are analyzed and resolved. The conclusion provides a useful reference for deeply understanding and reasonably using PCA-based face recognition algorithms.

关 键 词：算法评价 主分量分析(PCA) 人脸识别 假设检验 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]