基于切割子模块的单样本人脸识别

One sample per person face recognition based on partitioned sub-blocks

针对单样本人脸识别问题,本文提出了一种基于单样本切割的子模块主成分分析方法。该方法将单样本人脸图片切割成大小相同、互不重叠的多个子模块,切割后的子模块集构成新的样本集。对所有子模块作主成分分析(PCA)并提取特征,同一人脸的子模块特征系数作为分类识别的依据。在ORL人脸库上的测试结果表明,同PCA,(PC)2A,Sub-pattern LDA相比,该方法具有更好的识别率。

In order to deal with the problem of face recognition with one sample per person, a method called sub-block Principal Component Analysis （PCA） based on partitions of the sample is presented in this paper. The method first divides the sample into a few sub-blocks which have equal size and are non-overlapping, and then treats all of the sub-blocks as a new sample set. Finally, PCA is performed on all the sub-blocks so as to extract features. Classification is done according to the projection coefficients of sub-blocks of a person. The proposed approach is implemented on the ORL database and outperforms other methods such as PCA, （PC）2A and Sub-pattern LDA.