Median Fisher Discriminator: a robust feature extraction method with applications to biometrics

YANG Jian1, YANG Jingyu1, ZHANG David2

Front. Comput. Sci. ›› 0

PDF(419 KB)
PDF(419 KB)
Front. Comput. Sci. ›› DOI: 10.1007/s11704-008-0029-4

Median Fisher Discriminator: a robust feature extraction method with applications to biometrics

  • YANG Jian1, YANG Jingyu1, ZHANG David2
Author information +
History +

Abstract

In existing Linear Discriminant Analysis (LDA) models, the class population mean is always estimated by the class sample average. In small sample size problems, such as face and palm recognition, however, the class sample average does not suffice to provide an accurate estimate of the class population mean based on a few of the given samples, particularly when there are outliers in the training set. To overcome this weakness, the class median vector is used to estimate the class population mean in LDA modeling. The class median vector has two advantages over the class sample average: (1) the class median (image) vector preserves useful details in the sample images, and (2) the class median vector is robust to outliers that exist in the training sample set. In addition, a weighting mechanism is adopted to refine the characterization of the within-class scatter so as to further improve the robustness of the proposed model. The proposed Median Fisher Discriminator (MFD) method was evaluated using the Yale and the AR face image databases and the PolyU(Polytechnic University) palmprint database. The experimental results demonstrated the robustness and effectiveness of the proposed method.

Cite this article

Download citation ▾
YANG Jian, YANG Jingyu, ZHANG David. Median Fisher Discriminator: a robust feature extraction method with applications to biometrics. Front. Comput. Sci., https://doi.org/10.1007/s11704-008-0029-4

References

1. Webb A . StatisticalPattern Recognition. London: Hodder Arnold, 1999 2. Turk M, Pentland A . Eigenfaces for recognition. Journal of Cognitive Neuroscience, 1991, 3(1): 71–86. doi:10.1162/jocn.1991.3.1.71 3. Lu G, Zhang D, Wang K . Palmprint recognition using eigenpalms features. Pattern Recognition Letters, 2003, 24(9–10): 1463–1467. doi:10.1016/S0167‐8655(02)00386‐0 4. Liu K, Cheng Y -Q, Yang J -Y, et al.. An efficient algorithm for Foley-Sammon optimalset of discriminant vectors by algebraic method. International Journal of Pattern Recognition and Artificial Intelligence, 1992, 6(5): 817–829. doi:10.1142/S0218001492000412 5. Swets D L, Weng J . Using discriminant eigenfeaturesfor image retrieval. IEEE Transactionson Pattern Analysis and Machine Intelligence, 1996, 18(8): 831–836. doi:10.1109/34.531802 6. Belhumeur P N, Hespanha J P, Kriengman D J . Eigenfaces vs. fisherfaces: recognition using class specificlinear projection. IEEE Transactions onPattern Analysis and Machine Intelligence, 1997, 19(7): 711–720. doi:10.1109/34.598228 7. Chen L F, Liao H -Y M, Lin J C, et al.. A new LDA-based face recognition system whichcan solve the small sample size problem. Pattern Recognition, 2000, 33(10): 1713–1726. doi:10.1016/S0031‐3203(99)00139‐9 8. Jin Z, Yang J Y, Hu Z S, et al.. Face recognition based on uncorrelated discriminanttransformation. Pattern Recognition, 2001, 34(7): 1405–1416. doi:10.1016/S0031‐3203(00)00084‐4 9. Yu H, Yang J . A direct LDA algorithm forhigh-dimensional data-with application to face recognition. Pattern Recognition, 2001, 34(10): 2067–2070. doi:10.1016/S0031‐3203(00)00162‐X 10. Yang J, Yang J Y, Why can LDA be performedin PCA transformed space? Pattern Recognition, 2003, 36(2): 563–566. doi:10.1016/S0031‐3203(02)00048‐1 11. Liu C J, Wechsler H . Robust coding schemes forindexing and retrieval from large face databases. IEEE Transactions on Image Processing, 2000, 9(1): 132–137. doi:10.1109/83.817604 12. Kim T -K, Kittler J . Locally linear discriminantanalysis technique for multi-modally distributed classes for facerecognition with a single model image. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(3): 318–327. doi:10.1109/TPAMI.2005.58 13. Wu X, Zhang D, Wang K . Fisherpalms based palmprint recognition. Pattern Recognition Letters, 2003, 24(15): 2829–2838. doi:10.1016/S0167‐8655(03)00141‐7 14. Yang M H . Kernel eigenfaces vs. kernel fisherfaces: face recognition usingkernel methods. In: : Proceed-ings of theFifth IEEE International Conference on Automatic Face and GestureRecognition. Washington D. C., 2002 : 215–220 15. Lu J, Plataniotis K N, Venetsanopoulos A N . Face recognition using kernel directdiscriminant analysis algorithms. IEEETransactions on Neural Networks, 2003, 14(1): 117–126. doi:10.1109/TNN.2002.806629 16. Yang J, Frangi A, Yang J Y, et al.. KPCA plus LDA: a complete kernel Fisher discriminantframework for feature extraction and recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(2): 230–244. doi:10.1109/TPAMI.2005.33 17. Loog M, Duin R P W, Haeb-Umbach R . Multiclass linear dimension reduction by weighted pairwiseFisher criteria. IEEE Transactions on PatternAnalysis and Machine Intelligence, 2001, 23(7): 762–766. doi:10.1109/34.935849 18. Koren Y, Carmel L . Robust linear dimensionalityreduction, IEEE Transactions on Visualizationand Computer Graphics, 2004, 10(4): 459–470. doi:10.1109/TVCG.2004.17 19. He X, Yan S, Hu Y, et al.. Face recognition using Laplacianfaces. IEEE Transactions on Pattern Analysis and MachineIntelligence, 2005, 27(3): 328–340. doi:10.1109/TPAMI.2005.55 20. Kwak K-C, Pedrycz W . Face recognition using afuzzy fisherface classifier. Pattern Recognition, 2005, 38(10): 1717–1732. doi:10.1016/j.patcog.2005.01.018 21. Fidler S, Skocaj D, Leonardis A . Combining reconstructive and discriminative subspacemethods for robust classification and regression by subsampling. IEEE Transactions on Pattern Analysis and MachineIntelligence, 2005, 27(3): 328–340. doi:10.1109/TPAMI.2005.55 22. Chen S C, Liu J, Zhou Z -H, Making FLDA applicable to face recognition with one sampleper person. Pattern Recognition, 2004, 37(7): 1553–1555. doi:10.1016/j.patcog.2003.12.010 23. Huang J, Yuen P C, Chen W S, et al.. Component-based LDA method for face recognitionwith one training sample. AMFG (Analysisand Modeling of Faces and Gestures), 2003, 120–126 24. Hawkins D M, McLachlan G J . High-Breakdown linear discriminantanalysis. Journal of the American StatisticalAssociation, 1997, 92: 136–143. doi:10.2307/2291457 25. He X, Fung W K . High breakdown estimationfor multiple populations with applications to discriminant analysis. Journal of Multivariate Analysis, 2000, 72(2): 151–162. doi:10.1006/jmva.1999.1857 26. Hubert M, Driessen K V . Fast and robust discriminantanalysis. Computational Statistics and Data Analysis, 2003, 45: 301–320. doi:10.1016/S0167‐9473(02)00299‐2 27. Croux C, Dehon C . Robust linear discriminantanalysis using S-estimators. Canadian Journalof Statistics, 2001, 29: 473–493. doi:10.2307/3316042 28. Croux C, Dehon C, Rousseeuw P J, et al.. Robust estimation of the conditional medianfunction at elliptical models. Statistics &Probability Letters, 2001, 51: 361–368. doi:10.1016/S0167‐7152(00)00176‐0 29. Rousseeuw P S, Driessen K V . A fast algorithm for theminimum covariance determinant estimator. Technometrics, 1999 (41): 212–223 30. Grimmett G R, Stirzaker D R . Probability and Random Processes. 2nd ed . Oxford: Clarendon Press, 1992 31. Wikipedia. . Thefree encyclopedia. 32. Marion A . AnIntroduction to Image Processing. London:Chapman and Hall, 1991 33. Yale face database. 34. Martinez A M, Benavente R . The AR face database. 35. Martinez A M, Benavente R . The AR face database. Computer Vision Center Technical Report #24, 1998 36. Zhang D D . Palmprint Authentication. Berlin: Springer, 2004
AI Summary AI Mindmap
PDF(419 KB)

Accesses

Citations

Detail

Sections
Recommended

/