[1]YANG Jing-yu,ZHENG Yu-jie.Discriminant dimensionality reduction based on QR decomposition and its application in face recognition[J].CAAI Transactions on Intelligent Systems,2007,2(6):48-53.
Copy

Discriminant dimensionality reduction based on QR  decomposition and its application in face recognition

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 2 Number of periods: 2007 6 Page number: 48-53 Column: 学术论文—机器感知与模式识别 Public date: 2007-12-25
Title:
Discriminant dimensionality reduction based on QR  decomposition and its application in face recognition
Author(s):
YANG Jing-yu; ZHENG Yu-jie
Department of Computer Science, Nanjing University of Science & Technology, Nan jing 210094，China
Keywords:
discriminant dimensionality reduction;  pattern recognition;  QR decomposition;  di rect linear discriminant analysis (DLDA)
CLC:
TP391.4
DOI:
-
Abstract:
Dimensionality reduction is an important research topic in pattern recognition. At present, partial dimensionality reduction methods lack an effective discrimin ant information preservation mechanism. And the small sample size problem often occurs when the Fisher discriminant criterion is used. In this paper, discrimina nt information preservation is briefly presented and a new direct lnear discrim inant analysis (DLDA) method, the DLDA/QR algorithm, is suggested. With this algo rithm, the objective function is optimized through QR decomposition, and then th e ef fective discriminant information is extracted from a smaller space. Experimental results from the ORL face database demonstrate the effectiveness of the propose d method. 
References:
［1］ FUKUNAGA K. Introduction to statistical pattern recognition［M］. New York : Academic Press, 1990.
［2］DUDA R O,HART P E,STORK D G. Pattern classification［M］. 2nd ed. New Yo rk:John Wiley & Sons, 2001.
［3］BELHUMEUR P N,HESPANHA J P,KRIEGMAN D J. Eigenfaces vs fisherfaces: recogn it ion using class specific linear projection［J］. IEEE Trans Pattern Analysis an d Machine Intelligence, 1997, 19(7):711-720.
［4］COMON P. Independent component analysis—a new concept［J］.Signal Process ing, 1994, 36(3):287-314.
［5］BARTLETT M S, MOVELLAN J R, SEJNOWSKI T J. Face recognition by independent c omponent analysis［J］. IEEE Transaction on Neural Network, 2002, 13(6):1450-146 4.
［6］HYVARINEN A. Fast and robust fixedpoint algorithm for independent compon ent analysis［J］. IEEE Transaction on Neural Network, 1999, 10(3):626-634.
［7］YANG Jian, ZHANG David, YANG Jingyu. Is ICA significantly better than PCA for face recognition［A］. ICCV 2005［C］.［S.l.］，2005.
［8］SCHOLKOPF B,SMOLA A,MULLER K R, et al. Nonlinear component analysis as a k ernel eigenvalue problem［J］. Neural Computation. 1998, 10(5):1299-1319.
［9］SCHOLKOPF B,MIKA S,BURGES C. Input space vs feature space in kernelbase d methods［J］. IEEE Trans on Neural Networks， 1999, 10(5):1000-1017.
［10］SMOLA A J, MIKA S,SCHOLKOPF B, et al. Regularized principal man ifolds［J］. Journal of Machine Learning Research, 2001, 1(3):179-209.［11］STANFORD D C,RAGTERY A E. Fingding curvilinear features in spatial point pa tterns: principal curve clustering with noise［J］. IEEE Trans on Pattern Analy sis and Machine Intelligence, 2000, 22(6):601-609. ［12］BENGIO Y,PAIEMENT J F,VINCENT P. Out of sample extensions for LLE, isomap , MDS, eigenmaps, and spectral clustering［A］. Neural Information Processing Syst ems 16 (NIPS’2003)［C］.［S.l.］，2003.【13］ROWEIS S T,SAUL L K. Nonlinear dimensionality reduction by locally linear embedding［J］. Science, 2000, 290:2323-2326.
［14］HE X,YAN S,HU Y, et al. Face recognition using Laplacian faces［J］. IEEE Trans Pattern and Machine Intell, 2005, 27(3):328-340.
［15］YANG J,YU H,KUNZ W. An Efficient LDA algorithm for face recognition［A］. Int ernational Conference on Automation, Robotics, and Computer Vision (ICARCV’2000) ［C］. Singapore, 2000.
［16］CHEN L F,LIAO Y M,LIN J C, et al.A new LDAbased face recognition system which can solve the small sample size problem［J］. Pattern Recognition, 2000, 33(10):1713-1726.
［17］FISHER R A. The use of multiple measurements in taxonomic problems［J］. Annals of Eugenics, 1936(7):178-188.
［18］WILKS S S.Mathematical statistics［M］. New York: Wiley,1962.
［19］DUDA R,HART P.Pattern classification and scene analysis［M］. New York: W iley,1973.
［20］YE J P,LI Q. A twostage linear discriminant analysis via QRdecomposit ion［J］. IEEE Trans Pattern Anal Machine Intell, 2005, 2(6):929-941.
［21］GOLUB G H,VAN LOAN C F.Matrix computations［M］. 3rd ed. Baltimore: The Johns Hopkins Univ Press, 1996.
Similar References:

Memo

-

Last Update: 2009-05-08

Copyright © CAAI Transactions on Intelligent Systems