New Adaptive Linear Discriminante Analysis for Face Recognition with SVM

We have applied new accelerated algorithm for linear discriminate analysis (LDA) in face recognition with support vector machine. The new algorithm has the advantage of optimal selection of the step size. The gradient descent method and new algorithm has been implemented in software and evaluated on the Yale face database B. The eigenfaces of these approaches have been used to training a KNN. Recognition rate with new algorithm is compared with gradient.

Authors:

• Mehdi Ghayoumi

Keywords:

• lda
• adaptive
• svm
• face recognition.

References:

[1] L.C. Jain., U. Halici, I. Hayashi and S.B. Lee,"Intelligent Biometric
Techniques in Fingerprint and Face Recognition," CRC Press, 1999.
[2] Z. SunG. Bebis and R. Miller, "Object Detection Using Feature Subset
Selection," Pattern Recognition, Elsevier, vol. 37, no. 11, pp.2165-2176,
2004.
[3] C .Lee and J. Hong,m "Optimizing Feature Extraction for Multiclass
Cases," IEEE International Conference on Computational Cybernetics
and Simulations, pp.2545-2548, 1997.
[4] J. Mao and A.K.Jain, "Discriminant Analysis Neural Networks," IEEE
International Conference on Neural Networks, San Francisco, pp.300-
305, 1993.
[5] C. Chatterjee, Z. Kang and,V.P. Roychowdhury, "Algorithms for
Accelerated Convergence of Adaptive PCA," IEEE Transaction of Neural
Networks, vol. 11, no. 2,pp.338-355,2000.
[6] H. Abrishami Moghaddam and Kh. Amiri-Zadeh, "Fast Adaptive
Algorithms and Networks for Class-separability Features," Pattern
Recognition, vol. 36,pp.1695-1702, 2003.
[7] K Fukunaga, "Introduction to Statistical Pattern Recognition," Academic
Press, New York, 2nd edition, 1990.
[8] C. Chatterjee and V.P Roychowdhury, "On Self-Organizing Algorithm
and Networks for Class-separability Features," IEEE Transaction of
Neural Networks, vol. 8, no. 3,pp.663-678, 1997.
[9] Benveniste, A. M. Metivier and Priouret, P.:Adaptive Algorithms and
Stochastic Approximations, Springer, Berlin, (1990)
[10] L. Ljung, "Analysis of recursive stochastic algorithms," IEEE
Transaction of Automat. vol. 22, no. 4,pp.551-575, 1997.
[11] S. Georghiades, N. Belhumeur,and D. J. Kriegman,"From few to many:
illumination cone models for face recognition under variable lighting and
pose," IEEE Transaction of Pattern Anal, Machine Intell, pp.643-660,
2001.