PCA类内平均脸法在人脸识别中的应用研究

Study for Within-Class Average Face Method Based on PCA in Face Recognition

人脸识别是生物特征识别技术中一个非常活跃的课题,取得了很多研究成果。统计主元分析法(Prin-cipal Components Analysis,PCA)是人脸特征提取和识别的常用方法之一。结合传统PCA算法的特点,提出了一种用类内平均脸对类内样本进行规范化的方法。该方法有效地增加了类间样本的识别距离、有效地缩小了类内样本的识别距离,从而提高了人脸正确识别率。基于ORL人脸数据库的实验结果表明,该方法正确识别率达到98%,在人脸识别的实际应用中是一种可行的方法。

Face recognition is an active subject in the area of biometrical recognition technology, and lots of achievements have been obtained. Principal Components Analysis （PCA） is a basic method widely used in face feature extraction and recognition. In this paper, combined with the characteristics of traditional PCA, a method based on normalization of within-class average face image is presented, in which the classification distance of between-class samples is enlarged, while the classification distance of within-class samples is reduced. Thus face correct recognition rate is improved. Experimental results on ORL face database show that the method discussed has reached 98% of correct recognition rate, and is feasible in practical applications of faee reeognition.