摘要
人脸识别是当前人工智能和模式识别的研究热点。二维独立分量分析(two-dimensional independentcomponent analysis,2DICA)是人脸特征描述和识别地一种非常有效的方法,但是必须有一定数量和代表性的训练样本的支持。当仅有一个训练样本时,该方法中的协方差矩阵就变成了零矩阵,方法就会失效。针对这一问题,提出了一种基于采样二维独立分量分析(sampled two-dimensional independent component analysis,S2DICA)人脸识别方法。该方法是在2DICA运算之前,首先对单训练样本进行采样,通过多频率采样可以获取多个不同频率下的采样样本,然后对采样样本进行2DICA特征提取,最后采用神经网络分类识别,对人脸库ORL和YEL作了相关实验,将该方法与GREY、PCA、ICA、2DICA、PC PCA、FLDA、Sampled FLDA等传统方法作了比较,最终证明了该方法可以有效地解决单训练样本人脸识别的问题。
Face recognition is an active research area in the artificial intelligence. Two-dimensional independent component analysis is an efficient method of face features descriptions and recognition, but it must depend on a lot of representative raining samples. The method will become invalidation and the scatter matrix will become a zero matrix when any class or person has only one training pattern available. This paper proposed a face recognition algorithm using sampled two-dimensional independent component analysis based on this problem. The method obtained multiple training samples from a single face image by multi channel sampling before two-dimensional independent component analysis transformation. Experimental results on the ORL face database and YALE face database show that the proposed method is feasible and has higher recognition performance compared with GREY, PCA, ICA, 2DICA, projection combined PCA, FLDA, sampled FLDA and other algorithms where only one sample image per person is available for training.
出处
《计算机应用研究》
CSCD
北大核心
2010年第1期345-347,共3页
Application Research of Computers
