广义主分量分析及人脸识别

传统的主分量分析和Fisher线性鉴别分析在处理图像识别问题时都是基于图像向量的。该文提出了一种直接基于图像矩阵的主分量分析方法,它的突出优点是大大加快了特征抽取的速度。在ORL标准人脸库上的试验结果表明,该文所提出的方法不仅在识别性能上优于传统的主分量分析方法和Fisher线性鉴别分析方法,而且特征抽取的速度得到了很大的提高。

Generalized Principal Component Analysis and Face Recognition

The conventional principal component analysis(PCA)and Fisher linear discriminant analysis(LDA)are both based on vectors.That is to say,as they are used for image feature extraction,the first task is to convert the image matrix into image vector.Generally,the high dimensional image vectors make the process of feature extraction using PCA or LDA very time-consuming,to overcome the drawback of PCA or LDA,in this paper,a novel PCA technique directly based on original image matrices is developed for image feature extraction.Experimental results on ORL face database show the proposed GPCA is more powerful and efficient than conventional PCA and LDA.