Abstract
The quadratic discriminant classifier (QDC) is a well-known parametric Bayesian classifier that has been successfully applied to statistical pattern recognition problems. One such application is in automatic face recognition where the number of training images per subject is often found to be much less than the length of the facial features. In such a case, the QDC cannot be used because the class-specific covariance matrix on which it depends is either poorly estimated or singular thereby resulting in unacceptable classifier performance. High dimensional covariance estimation techniques such as shrinkage can alleviate this problem but only to a certain extent. This paper presents a computationally simple yet effective solution for further improving the QDC performance in small sample size scenarios. The proposed technique adopts a strategy of combining the class-specific shrinkage estimates of the covariance matrix to obtain a pooled shrinkage estimate, which is then plugged into the QDC. Experiments indicate that the proposed classifier leads to remarkable improvement in face recognition accuracy as compared to the existing classifiers such as the nearest neighbor, support vector machine and naive Bayes, irrespective of the nature of the database and feature extraction method. Monte Carlo simulations reveal that this improvement is due to the much lower mean squared error of the pooled shrinkage estimator which offers greater stability to the QDC.
Similar content being viewed by others
References
Ahonen T, Hadid A (2006) Face description with local binary patterns: application to face recognition. IEEE Trans Pattern Anal Mach Intell 28(12):2037–2041
Ali SS, Howlader T, Rahman SMM (2014) Quadratic discriminant classifier for Gauss-Hermite moment-based face recognition. In: Proceedings of the ISI regional statistical conference, pp 1–12. Kuala Lumpur, Malaysia, Nov 16–19 2014. International Statistical Institute
Bai J, Shi S (2011) Estimating high dimensional covariance matrices and its applications. Ann Econ Fin 12(2):199–215
Belhumeur PN, Hespanha JP, Kreigman DJ (1997) Eigenfaces vs. Fisherfaces: recognition using class specific linear projection. IEEE Trans Pattern Anal Mach Intell 19(7):711–720
Chen BC, Chen CS, Hsu WH (2014) Cross-age reference coding for age-invariant face recognition and retrieval. In: Proceedings of the European conference on computer vision, pp 768–783. Zurich, Switzerland
Chen Y, Wiesel A, Eldar YC, Hero AO (2010) Shrinkage algorithms for MMSE covariance estimation. IEEE Trans Signal Process 58(10):5016–5029
DeMiguel V, Martin-Utrera A (2013) Size matters: optimal calibration of shrinkage estimators for portfolio selection. J Bank Finan 37(8):3018–3034
Friedman JH (1989) Regularized discriminant analysis. J Am Stat Assoc 84(405):165–175
Hoffbeck JP, Landgrebe DA (1996) Covariance matrix estimation and classification with limited training data. IEEE Trans Pattern Anal Mach Intell 18(7):763–767
Huang GB, Ramesh M, Berg T, Miller EL (2007) Labeled faces in the wild: a database for studying face recognition in unconstrained environments. Technical Report 07-49, University of Massachusetts, Amherst
Jain AK, Duin RPW, Mao J (2000) Statistical pattern recognition: a review. IEEE Trans Pattern Anal Mach Intell 22(1):4–37
Janzamin M, Anandkumar A (2014) High-dimensional covariance decomposition into sparse Markov and independence models. J Mach Learn Res 15(Apr):1549–1591
Kanade T, Cohn JF, Tian YL (2000) Comprehensive database for facial expression analysis. In: Proceedings of the IEEE international conference on automatic face and gesture recognition, pp 484–490. Grenoble, France, Mar 28–30, 2000. IEEE Biometrics Council
Kittler J (1994) Statistical pattern recognition in image analysis. J Appl Stat 21(1–2):61–75
Kwan CCY (2008) Estimation error in the average correlation of security returns and shrink- age estimation of covariance and correlation matrices. Finan Res Lett 5(1):236–244
Ledoit O, Wolf M (2003) Improved estimation of the covariance matrix of stock returns with an application to portfolio selection. J Emp Finan 10(5):1–20
Ledoit O, Wolf M (2004) A well-conditioned estimator for large-dimensional covariance matrices. J Multivar Anal 88:365–411
Ledoit O, Wolf M (2012) Nonlinear shrinkage estimation of large-dimensional covariance matrices. Ann Stat 40(2):1024–1060
Li M, Yuan B (2005) 2D-LDA: a statistical linear discriminant analysis for image matrix. Pattern Recogn Lett 26(5):527–532
Lu J, Plataniotis KN, Venetsanopoulos AN (2003) Regularized discriminant analysis for the small sample size problem in face recognition. Pattern Recogn Lett 24(16):3079–3087
Lu SX, Wang XZ, Zhang GQ, Zhou X (2015) Effective algorithms of the Moore-Penrose inverse matrices for extreme learning machine. Intell Data Anal 19(4):743–760
Mitra S, Lazar NA, Liu Y (2007) Understanding the role of facial asymmetry in human face identification. Stat Comput 17(1):57–70
Phillips PJ, Flynn PJ, Scruggs T, Bowyer KW, Chang J, Hoffman K, Marques J, Min J, Worek W (2005) Overview of the face recognition grand challenge. In: Proceedings of the IEEE international conference on computer vision and pattern recognition, pp 947–954, San Diego, CA, Jun 20–26, 2005. IEEE Computer Society
Pourahmadi M (2011) Covariance estimation: the GLM and regularization perspectives. Ann Econ Finan 26(3):369–387
Pourahmadi M (2013) High-dimensional covariance estimation: with high-dimensional data, 1st edn. Wiley, NJ
Rahman SMM, Lata SP, Howlader T (2015) Bayesian face recognition using 2D Gaussian-Hermite moments. EURASIP J Image Video Process 2015(35):1–20
Rivera AR, Castillo JR, Chae O (2013) Local directional number pattern for face analysis: face and expression recognition. IEEE Trans Image Process 22(5):1740–1752
Schafer J, Strimmer K (2005) A shrinkage approach to large-scale covariance matrix estimation and implications for functional genomics. Stat Appl Genet Mol Biol 4(1):1544–6115
Sharma A, Paliwal KK (2015) Linear discriminant analysis for the small sample size problem: an overview. Int J Mach Learn Cybern 6(3):443–454
Shen J, Shen W, Shen D (2000) On geometric and orthogonal moments. Int J Pattern Recogn Artif Intell 14(7):875–894
Srivastava S, Gupta MR, Frigyik BA (2007) Bayesian quadratic discriminant analysis. J Mach Learn Res 8(Jun):1277–1305
Stein C (1975) Estimation of a covariance matrix. In: Rietz Lecture, Atlanta, GA, 1975. 39th annual meeting
Sun Y, Chen Y, Wang X, Tang X (2014) Deep learning face representation by joint identification-verification. In: Proceedings of advances in neural information processing systems 27, pp 1988–1996, Montreal, Canada, Dec 8–13, 2014. Curran Associates, Inc
Thomaz CE, Gillies DF, Feitosa RQ (2004) A new covariance estimate for Bayesian classifiers in biometric recognition. IEEE Trans Circuits Syst Video Technol 14(2):214–223
Tian YL, Kanade T, Cohn JF (2001) Recognizing action units for facial expression analysis. IEEE Trans Pattern Anal Mach Intell 23(2):97–115
Turk M, Pentland A (1991) Eigenfaces for recognition. J Cogn Neurosci 3(1):71–86
Viola P, Jones MJ (2004) Robust real-time face detection. Int J Comput Vis 57(2):137–154
Wang C, Pan G, Tong T, Zhu L (2015a) Shrinkage estimation of large dimensional precision matrix using random matrix theory. Stat Sin 25(3):993–1008
Wang XZ (2015) Learning from big data with uncertainty editorial. J Intell Fuzzy Syst 28(5):2329–2330
Wang XZ, Ashfaq RAR, Fu AM (2015b) Fuzziness based sample categorization for classifier performance improvement. J Intell Fuzzy Syst 29(3):1185–1196
Won J, Lim J, Kim S, Rajaratnam B (2013) Condition-number-regularized covariance estimation. J R Stat Soc B 75(3):427–450
Wright J, Yang AY, Ganesh A, Sastry SS, Ma Y (2009) Robust face recognition via sparse representation. IEEE Trans Pattern Anal Mach Intell 31(2):210–227
Yang B, Dai M (2011) Image analysis by Gaussian-Hermite moments. Signal Process 91(10):2290–2303
Yang B, Li G, Zhang H, Dai M (2011) Rotation and translation invariants of Gaussian-Hermite moments. Pattern Recogn Lett 32(9):1283–1298
Yang J, Zhang D, Frangi AF, Yang J-Y (2004) Two-dimensional PCA: a new approach of appearance-based face representation and recognition. IEEE Trans Pattern Anal Mach Intell 26(1):131–137
Yang M, Zhu PF, Liu F, Shen LL (2015) Joint representation and pattern learning for robust face recognition. Neurocomputing 168:70–80
Yao J, Chang C, Salmi ML, Hung YS, Loraine A, Roux SJ (2008) Genome-scale cluster analysis of replicated microarrays using shrinkage correlation coefficient. BMC Bioinf 9(288):1–16
Acknowledgments
The authors would like to give thanks to the anonymous reviewers for their valuable comments that were useful to improve the quality of the paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ali, S.S., Howlader, T. & Rahman, S.M.M. Pooled shrinkage estimator for quadratic discriminant classifier: an analysis for small sample sizes in face recognition. Int. J. Mach. Learn. & Cyber. 9, 507–522 (2018). https://doi.org/10.1007/s13042-016-0549-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13042-016-0549-4