Abstract
This paper proposes a new face recognition method that combines the ensemble learning with the third-order Gabor tensor. In this method, the third-order Gabor tensor is used to replace the vectorial Gabor feature representation in order to keep high-dimensional adjacent structures in images. In order to avoid to fall into the curse of the dimensions due to the tensor, a multilinear principle component analysis (MPCA) algorithm is utilized to reduce the dimensions of the Gabor tensor. The obtained low-dimensional Gabor tensor features are selected in term of their discriminant ability to form a vectorial Gabor feature representation. It is embedded into a new sample selection scheme to construct a new classifier. Different from the traditional sample selection, the samples with high misclassification rate regardless of their class is used to train a set of diversity Nonparametric Maximum Margin Criterion (NMMC) learners and the scheme allows each class to have different numbers of samples. In construction of the classifier, multiple weak classifiers are first trained in terms of the K-NN criterion and then these weak classifiers are fused into a boosted classifier in terms of the confidence levels of individual weak classifiers. The proposed method inherits the merit of both the boosting technique and the Gabor wavelets. Experimental results on several benchmark face databases show that it attains better performance than the existing state-of-the-art methods.
Similar content being viewed by others
References
Chen LF, Liao HYM, Ko MT, Lin JC, Yu GJ (2000) A new LDA-based face recognition system which can solve the small sample size problem. Pattern Recogn 33(10):1713–1726
Deypir M, Alizadeh S, Zoughi T, Boostani R (2011) Boosting a multi-linear classifier with application to visual lip reading. Expert Syst Appl 38(1):941–948
FERET database. From http://mloss.org/software/view/416/
Kearns MJ, Valiant LG (2003) The boosting approach to machine learning: An overview. Nonlinear Estimation and Classification. Springer, New York, pp 149–171
Liao P, Liu J, Wang M, Ma H, Zhang W (2012) Ensemble local fractional LDA for face recognition. Proceedings of IEEE International Conference on Computer Science and Automation Engineering (CSAE) 3:586–590
Liu C, Wechsler H (2002) Gabor feature based classification using the enhanced fisher linear discriminant model for face recognition. IEEE Trans Image Process 11(4):467–476
Lu J, Plataniotis KN, Venetsanopoulos AN, Li SZ (2006) Ensemble-based discriminant learning with boosting for face recognition. IEEE Trans Neural Netw 17(1):166–178
Lu H, Plataniotis KN, Venetsanopoulos AN (2008) MPCA: multilinear principal component analysis of tensor objects. IEEE Trans Neural Netw 19(1):18–39
Lu H, Plataniotis KN, Venetsanopoulos AN (2009) Boosting discriminant learners for gait recognition using MPCA features. Journal on Image and Video Processing 2009(1):1–11
Olshausen BA (1996) Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381(6583):607–609
ORL Database . From http://www.cad.zju.edu.cn/home/dengcai/Data/FaceData.html
Qiu X, Wu L (2005) Nonparametric maximum margin criterion for face recognition. Proceedings of IEEE international conference on image processing 2:II-918-21
Schapire RE, Freund Y, Bartlett P, Lee WS (1998) Boosting the margin: a new explanation for the effectiveness of voting methods. Ann Stat 26(5):1651–1686
Turk M, Pentland AP (1991) Face recognition using eigenfaces. Proceedings of IEEE computer society conference on computer vision and pattern recognition. CVPR 1991. doi:10.1109/CVPR.1991.139758, pp 586–591
Wang X, Tang X (2004) Random sampling LDA for face recognition[C]. Computer Vision and Pattern Recognition, CVPR, Proceedings of the IEEE computer society conference on. IEEE, 2: II-259-II-265 Vol. 2
Wold S, Esbensen K, Geladi P (1987) Principal component analysis. Chemom Intell Lab Syst 2(1):37–52
Wu J, Nian X, Yang W, Sun C (2015) MPCA on Gabor tensor for face recognition. Proceedings of the 2015 Chinese Intelligent Automation Conference Lecture Notes in Electrical Engineering 336:421–429
XM2VTS database. From http://www.ee.surrey.ac.uk/CVSSP/xm2vtsdb/
Yu H, Yang J (2001) A direct LDA algorithm for high-dimensional data—with application to face recognition. Pattern Recogn 34(10):2067–2070
Zhao W, Chellappa R, Phillips PJ, Rosenfeld A (2003) Face recognition: a literature survey. ACM Comput Surv (CSUR) 35(4):399–458
Acknowledgments
This research was supported in part by the National Natural Science Foundation of China (Grant No. 61101246) and the Fundamental Research Funds for the Central Universities (Grant No. JB150209).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Liu, J., A., P., Ge, Q. et al. Gabor tensor based face recognition using the boosted nonparametric maximum margin criterion. Multimed Tools Appl 77, 9055–9069 (2018). https://doi.org/10.1007/s11042-017-4805-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-017-4805-8