分块PCA及其在人脸识别中的应用

Modular PCA and its application in human face recognition

主成分分析(principal component analysis,PCA)是公认的特征抽取的最为重要的工具之一,目前仍然被广泛地应用在人脸等图像识别领域。基于PCA,提出了分块PCA的人脸识别方法。分块PCA方法先对图像进行分块,对分块得到的子图像利用PCA进行鉴别分析。其特点是能有效地抽取图像的局部特征,对人脸表情和光照条件变化较大的图像表现尤为突出。与PCA方法相比,由于使用子图像矩阵,分块PCA可以避免使用奇异值分解理论,过程简便。此外,PCA是分块PCA的特例。在Yale和NUST603人脸库上的试验结果表明,所提出的方法在识别性能上明显优于经典的PCA方法,识别率可以分别提高6.7和4.4个百分点。

Principal component analysis （PCA） is oneof accepted and important technique for feature extraction widely used in the areas of images recognition such as human face recognition. Modular PCA, a human face recognition technique based on PCA, is presented in this paper. First, in proposed approach, the original images are divided into modular images, which are also called sub- images. Then, the well-known PCA method is directly used to the sub-images obtained from the previous step. There are two advantages for this way： 0）local feature of the images can be extracted efficiently, and it is really true of the images that have large variations in facial expression and lighting; ~singnlar value decomposition of matrix may be avoided in the process of feature extraction, which is simple than that of other technologies such as PCA. Moreover, PCA is a special case of modular PCA. To test modular PCA and to evaluate its performance, a series of experiments are performed on two human face image databases： Yale and NJUST603 human face databases. The experimental results indicate that the performance of modular PCA is obviously superior to that of PCA. The recognition rates, which may be obtained, are 100 percent and 97.5 percent with respect to the two databases, respectively.