Single Sample Face Recognition with Gabor Feature based Linear Regression

摘要：

　　By constructing a linear model representing using the auxiliary intra-personal variations,the adaptive linear regression classifier(ALRC)successfully popularized the linear regression classifier(LRC)to the single sample per person(SSPP)scenario.However,ALRC simply uses original face images to constitute the feature space,which is not robust enough against the variations of probe images,and leads to the algorithm computationally very expensive.To address the two problems,in this paper,the image Gabor features is used for ALRC scheme.By using Gabor kernels based feature space,the abilities of ALRC against variations can be improved.Meanwhile,principal component analysis(PCA)is implemented in the feature extracted stage.Experiments on representative face databases show that Gabor-feature based ALRC can achieve better recognition performance than original ALRC,and reduce much computational burden compared with later.

关键词： Face Recognition LRC ALRC Gabor wavelets PCA

作者: CHEN Xin ZHANG Hongbin

作者单位: School of Electronic Engineering,University of Electronic Science and Technology of China,Chengdu,Si School of Electronic Engineering,University of Electronic Science and Technology of China,Chengdu,Si

会议类型: 国际会议

会议名称: The 33th Chinese Control Conference第33届中国控制会议

会议地点: 南京

会议语种:英文

页码: 4910-4913

在线出版日期: 2014-07-28（万方平台首次上网日期，不代表论文的发表时间）