点击次数：

论文类型：期刊论文

发表时间：2015-11-01

发表刊物：Journal of Computational and Theoretical Nanoscience

收录刊物：EI

卷号：12

期号：11

页面范围：4873-4880

ISSN号：15461955

摘要：To extract effective and reliable characteristic is the emphasis of research on the human recognition. In this paper, firstly, by using principal component analysis (PCA) to reduce the dimension of feature vector, so as not only to ensure the dimension of the feature data as low as possible, but also as far as possible to maintain the information difference retaining in the original data. Next, the local binary pattern (LBP) widely used in texture recognition has been introduced to face recognition, for face classification less variety, LBP and its derivative operator demonstrates better recognition rate. In order to reflect the structural characteristics of human face, this paper has also carried out the block processing, and through this procedure its experimental recognition rate is obviously better than the recognition rate of that untreated. ? 2015 American Scientific Publishers.