A face recognition algorithm using a fusion method based on Adaboost Bidirectional 2DLDA

A challenge for face recognition is variation, such as due to lighting or facial expression differences. To solve this problem, we fuses bidirectional two-dimensional linear discriminant analysis (2DLDA) feature by adaboost technique and propose a novel recognition method called AB2DLDA in this paper. This method can perform well with small number of samples. In this paper, firstly we analyze complementarity for vertical direction of 2DLDA and horizontal direction of E2DLDA. Then we use adaboost to design a classifier, which improves recognition performance by fusing 2DLDA and E2DLDA. Finally, our method is tested on AR face databases. Experimental results show that our method functions with good recognition accuracy and robustness.

[1]  J. Friedman.  Special Invited Paper-Additive logistic regression: A statistical view of boosting , 2000 .

[2]  Hong Yan,et al.  Unified formulation of linear discriminant analysis methods and optimal parameter selection , 2011, Pattern Recognit..

[3]  Jieping Ye,et al.  A two-stage linear discriminant analysis via QR-decomposition , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Jin-gao Liu,et al.  Gabor-2DLDA: Face Recognition Using Gabor Features and 2D Linear Discriminant Analysis , 2009, 2009 Second International Conference on Intelligent Computation Technology and Automation.

[5]  Yong Wang,et al.  Incremental complete LDA for face recognition , 2012, Pattern Recognit..

[6]  Jun Liu,et al.  Discriminant common vectors versus neighbourhood components analysis and Laplacianfaces: A comparative study in small sample size problem , 2006, Image Vis. Comput..

[7]  Fangjun Kuang,et al.  Complete Fuzzy LDA Algorithm in Image Segmentation , 2012 .

[8]  Chris H. Q. Ding,et al.  Symmetric two dimensional linear discriminant analysis (2DLDA) , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Ja-Chen Lin,et al.  A new LDA-based face recognition system which can solve the small sample size problem , 1998, Pattern Recognit..

[10]  Pong C. Yuen,et al.  Face Recognition by Regularized Discriminant Analysis , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[11]  M. Omair Ahmad,et al.  Optimizing the kernel in the empirical feature space , 2005, IEEE Transactions on Neural Networks.

[12]  Konstantinos N. Plataniotis,et al.  Ensemble-based discriminant learning with boosting for face recognition , 2006, IEEE Transactions on Neural Networks.

[13]  Aleix M. Martinez,et al.  The AR face database , 1998 .

[14]  Gao Zheng.  A Survey of Automatic Human Face Recognition , 2003 .

[15]  Xiaogang Wang,et al.  Random Sampling for Subspace Face Recognition , 2006, International Journal of Computer Vision.

[16]  Nitesh V. Chawla,et al.  Random subspaces and subsampling for 2-D face recognition , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[17]  Hanqing Lu,et al.  Solving the small sample size problem of LDA , 2002, Object recognition supported by user interaction for service robots.

[18]  Hanyang Jiang.  Research on the Document Image Segmentation Based on the LDA Model , 2012 .