Please wait a minute...
Frontiers of Computer Science

ISSN 2095-2228

ISSN 2095-2236(Online)

CN 10-1014/TP

Postal Subscription Code 80-970

2018 Impact Factor: 1.129

Front Comput Sci Chin    2011, Vol. 5 Issue (4) : 419-428    https://doi.org/10.1007/s11704-011-1012-z
RESEARCH ARTICLE
Nearest-neighbor classifier motivated marginal discriminant projections for face recognition
Pu HUANG1(), Zhenmin TANG1, Caikou CHEN2, Xintian CHENG1
1. School of Computer Science and Technology, Nanjing University of Science and Technology, Nanjing 210094, China; 2. College of Information Engineering, Yangzhou University, Yangzhou 225009, China
 Download: PDF(267 KB)   HTML
 Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Abstract

Marginal Fisher analysis (MFA) is a representative margin-based learning algorithm for face recognition. A major problem in MFA is how to select appropriate parameters, k1 and k2, to construct the respective intrinsic and penalty graphs. In this paper, we propose a novel method called nearest-neighbor (NN) classifier motivated marginal discriminant projections (NN-MDP). Motivated by the NN classifier, NN-MDP seeks a few projection vectors to prevent data samples from being wrongly categorized. Like MFA, NN-MDP can characterize the compactness and separability of samples simultaneously. Moreover, in contrast to MFA, NN-MDP can actively construct the intrinsic graph and penalty graph without unknown parameters. Experimental results on the ORL, Yale, and FERET face databases show that NN-MDP not only avoids the intractability, and high expense of neighborhood parameter selection, but is also more applicable to face recognition with NN classifier than other methods.

Keywords dimensionality reduction (DR)      face recognition      marginal Fisher analysis (MFA)      locality preserving projections (LPP)      graph construction      margin-based      nearest-neighbor (NN) classifier     
Corresponding Author(s): HUANG Pu,Email:huangpu3355@163.com   
Issue Date: 05 December 2011
 Cite this article:   
Pu HUANG,Zhenmin TANG,Caikou CHEN, et al. Nearest-neighbor classifier motivated marginal discriminant projections for face recognition[J]. Front Comput Sci Chin, 2011, 5(4): 419-428.
 URL:  
https://academic.hep.com.cn/fcs/EN/10.1007/s11704-011-1012-z
https://academic.hep.com.cn/fcs/EN/Y2011/V5/I4/419
Fig.1  Samples in original space
Fig.2  Samples after projection
Fig.3  10 images of one person in ORL database
Fig.4  11 images of one person in Yale database
Fig.5  7 images of one person in FERET database
Fig.6  Recognition accuracy of LPP corresponding to different with different classifiers
Fig.7  Recognition accuracy of MFA corresponding to different and with (a) NN; (b) SVM
AlgorithmsRecognition accuracy/%Dimensiontime/s
PCA-NN88.33641.8280
PCA-SVM89.17391.2190
LDA-NN90.42201.5780
LDA-SVM90.83321.5930
LPP-NN90.42 (k = 2)354.5160
LPP-SVM91.67 (k = 2)354.3590
MFA-NN92.50 (k1 = 1, k2 = 60)286.7030
MFA-SVM92.92 (k1 = 1, k2 = 80)316.8900
SMDA-NN85.833411.7500
SMDA-SVM86.673612.3590
NN-MDP-NN93.75211.7190
NN-MDP-SVM92.92201.6570
Tab.1  Recognition accuracy, dimension, and consuming time of algorithms (PCA, LDA, LPP, MFA, SMDA and NN-MDP) on ORL database
Recognition accuracy/%
AlgorithmsSample size=5Sample size=6Sample size=7
PCA-NN93.3392.00100
PCA-SVM91.1189.3396.67
LDA-NN97.7896.00100
LDA-SVM97.7896.00100
LPP-NN92.2294.6796.67
LPP-SVM96.6798.6798.33
MFA-NN97.7896.00100
MFA-SVM97.7896.00100
SMDA-NN97.7898.67100
SMDA-SVM98.89100100
NN-MDP-NN97.7897.13100
NN-MDP-SVM93.3393.33100
Tab.2  Recognition accuracy of algorithms (PCA, LDA, LPP, MFA, SMDA and NN-MDP) on Yale database
AlgorithmsRecognition accuracy/%Dimensiontime/s
PCA-NN45.25607.1560
PCA-SVM49.2511010.3900
LDA-NN52.502012.7660
LDA-SVM51.752013.2810
LPP-NN36.136047.7500
LPP-SVM40.0013050.6100
MFA-NN52.002052.3750
MFA-SVM51.882052.7030
SMDA-NN51.88140178.5940
SMDA-SVM53.25140179.7810
NN-MDP-NN52.755014.7030
NN-MDP-SVM52.386014.9380
Tab.3  Recognition accuracy, dimension, and consuming time of algorithms (PCA, LDA, LPP, MFA, SMDA and NN-MDP) on FERET database in Case 1
AlgorithmsRecognition accuracy/%DimensionConsuming time/s
PCA-NN59.50908.6410
PCA-SVM67.6314011.7190
LDA-NN73.004013.1410
LDA-SVM73.509013.9690
LPP-NN48.2511049.0630
LPP-SVM61.2514049.2660
MFA-NN73.753054.1250
MFA-SVM74.6311055.3130
SMDA-NN68.13140169.5460
SMDA-SVM78.50140180.0310
NN-MDP-NN76.006015.0780
NN-MDP-SVM77.006015.5310
Tab.4  Recognition accuracy, dimension, and consuming time of algorithms (PCA, LDA, LPP, MFA, SMDA and NN-MDP) on FERET database in Case 2
1 Turk M, Pentland A. Eigenfaces for recognition. Journal of Cognitive Neuroscience , 1991, 3(1): 71-86
2 Martinez A M, Kak A C. PCA versus LDA. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2001, 23(2): 228-233
3 Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. Fisherfaces: recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence , 1997, 19(7): 711-720
4 Tenenbaum J B, De S V, Langford J C. A global geometric framework for nonlinear dimensionality reduction. Science , 2000, 290(5500): 2319-2323
5 Roweis S T, Saul L K. Nonlinear dimensionality reduction by locally linear embedding. Science , 2000, 290(5500): 2323-2326
6 Belkin M, Niyogi P. Laplacian Eigenmaps for dimensionality reduction and data representation. Neural Computation , 2003, 15(6): 1373-1396
7 Yan S C, Xu D, Zhang B Y, Zhang H J, Yang Q, Lin S. Graph embedding and extensions: A general framework for dimensionality reduction. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2007, 29(1): 40-51
8 He X, Niyogi P. Locality preserving projections. In: Proceedings of 2003 Neural Information Processing Systems . 2003, 153-160
9 Yang J, Zhang D, Yang J Y, Niu B. Globally maximizing, locally minimizing: unsupervised discriminant projection with applications to face and palm biometrics. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2007, 29(4): 650-664
10 Yu W W, Teng X L, Liu C Q. Face recognition using discriminant locality preserving projections. Image and Vision Computing , 2006, 24(3): 239-248
11 Zhao H T, Sun S Y, Jing Z L, Yang J Y. Local structure based supervised feature extraction. Pattern Recognition , 2006, 39(8): 1546-1550
12 Sugiyama M. Dimensionality reduction of multimodal labeled data by local Fisher discriminant analysis. Journal of Machine Learning Research , 2007, 8(May): 1027-1061
13 Xu D, Yan S C, Tao D C, Lin S, Zhang H J. Marginal Fisher analysis and its variants for human Gait recognition and content-based image retrieval. IEEE Transactions on Image Processing , 2007, 16(11): 2811-2821
14 Huang P, Chen C K. Enhanced marginal Fisher analysis for face recognition. In: Proceedings of 2009 International Conference on Artificial Intelligence and Computational Intelligence . 2009, 403-407
15 Zhao C R, Lai Z H, Sui Y, Chen Y. Local maximal marginal embedding with application to face recognition. In: Proceeding of 2008 Chinese Conference on Pattern Recognition . 2008, 1-6
16 Xu J, Yang J. Nonparametric Marginal Fisher analysis for feature extraction. In: Proceedings of 6th International Conference on Intelligent Computing . 2010, 221-228
17 Cai D, He X, Zhou K, Han J, Bao H. Locality sensitive discriminant analysis. In: Proceedings of 20th International Joint Conference on Artificial Intelligence . 2007, 708-713
18 Yang B, Chen S C. Sample-dependent graph construction with application to dimensionality reduction. Neurocomputing , 2010, 74(1-3): 301-314
19 Zhang L, Qiao L, Chen S C. Graph-optimized locality preserving projections. Pattern Recognition , 2010, 43(6): 1993-2002
20 Qiao L, Chen S C. Sparsity preserving discriminant analysis for single training image face recognition. Pattern Recognition Letters , 2010, 31(5): 422-429
21 Qiao L, Chen S C, Tan X. Sparsity preserving projections with applications to face recognition. Pattern Recognition , 2010, 43(1): 331-341
22 Li S Z, Lu J W. Face recognition using the nearest feature line method. IEEE Transactions on Neural Networks , 1999, 10(2): 439-443
23 Yu H, Yang J. A direct LDA algorithm for high-dimensional data with application to face recognition. Pattern Recognition , 2001, 34(11): 2067-2070
24 Gu Z H, Yang J. Sparse margin based discriminant analysis for face recognition. In: Proceedings of 17th IEEE International Conference on Image Processing . 2010, 1669-1672
25 Huang J B, Yang M H. Fast sparse representation with prototypes. In: Proceedings of 23rd IEEE Conference on Vision and Pattern Recognition . 2010, 3618-3625
26 Vapnik V N. The Nature of Statistical Learning Theory. New York: Springer-Verlag, 1995
[1] Jin LI, Quan CHEN, Jingwen LENG, Weinan ZHANG, Minyi GUO. Probabilistic robust regression with adaptive weights–a case study on face recognition[J]. Front. Comput. Sci., 2020, 14(5): 145314-.
[2] Yan LI, Shiguang SHAN, Ruiping WANG, Zhen CUI, Xilin CHEN. Fusing magnitude and phase features with multiple face models for robust face recognition[J]. Front. Comput. Sci., 2018, 12(6): 1173-1191.
[3] Xin LIU,Meina KAN,Wanglong WU,Shiguang SHAN,Xilin CHEN. VIPLFaceNet: an open source deep face recognition SDK[J]. Front. Comput. Sci., 2017, 11(2): 208-218.
[4] Qicong WANG,Binbin WANG,Xinjie HAO,Lisheng CHEN,Jingmin CUI,Rongrong JI,Yunqi LEI. Face recognition by decision fusion of two-dimensional linear discriminant analysis and local binary pattern[J]. Front. Comput. Sci., 2016, 10(6): 1118-1129.
[5] Lishan QIAO, Limei ZHANG, Songcan CHEN. Dimensionality reduction with adaptive graph[J]. Front Comput Sci, 2013, 7(5): 745-753.
[6] YANG Jian, YANG Jingyu, ZHANG David. Median Fisher Discriminator: a robust feature extraction method with applications to biometrics[J]. Front. Comput. Sci., 2008, 2(3): 295-305.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed