Normalization Discriminant Independent Component Analysis

In face recognition, feature extraction techniques
attempts to search for appropriate representation of the data. However,
when the feature dimension is larger than the samples size, it brings
performance degradation. Hence, we propose a method called
Normalization Discriminant Independent Component Analysis
(NDICA). The input data will be regularized to obtain the most
reliable features from the data and processed using Independent
Component Analysis (ICA). The proposed method is evaluated on
three face databases, Olivetti Research Ltd (ORL), Face Recognition
Technology (FERET) and Face Recognition Grand Challenge
(FRGC). NDICA showed it effectiveness compared with other
unsupervised and supervised techniques.

• Liew Yee Ping
• Pang Ying Han
• Lau Siong Hoe
• Ooi Shih Yin
• Housam Khalifa Bashier Babiker

• Face recognition
• small sample size
• regularization
• independent component analysis.

<p>[1] M. A. Turk and A. P. Pentland, &ldquo;Face Recognition Using Eigenfaces&rdquo;,
IEEE Conf. on Computer Visionand Pattern Recognition, pp. 586-591,
1991
[2] Chelali, F.Z.; Djeradi, A.; Djeradi, R.,&quot;Linear discriminant analysis for
face recognition,&quot; Multimedia Computing and Systems, 2009. ICMCS &#39;09.
International Conference on , vol., no., pp.1,10, 2-4 April 2009
[3] M. S. Bartlett, H. M. Lades, and T. J. Sejnowski, &quot;Independent
component representations for face recognition,&quot; presented at SPIE
Symposium on Electronic Imaging: Science and Technology; Conference
on Human Vision and Electronic Imaging III, San Jose, CA, 1998.
[4] Ying-Han Pang, Andrew Teoh Beng Jin, David Ngo Chek Ling, &ldquo;Face
authentication system using pseudo Zernike moments on wavelet
subband&rdquo;. Proc. IEICE Electr. Exp. v1 i10. 275-280
[5] Wiskott, L.; Fellous, J.-M.; Kruger, Norbert; Von der Malsburg, C., &quot;Face
recognition by elastic bunch graph matching,&quot; Image Processing, 1997.
Proceedings., International Conference on , vol.1, no., pp.129,132 vol.1,
26-29 Oct 1997
[6] I. Sirovich and M. Kirby, &ldquo;Low- dimensional procedure for the
caracterization of human faces,&rdquo;Journal of Optical Society of America A,
vol. 4, no. 3, pp. 519&ndash;524, March 1987.
[7] Peter N. Belhumeur, J. P. Hespanha, David J. Kriegman, Eigenfaces vs.
Fishefaces: Recognition Using Class Specific Linear Projection, IEEE
Transactions on Pattern Analysis and Machine Intelligence, v.19 n.7,
p.711-720, July 1997.
[8] Shiladitya Chowdhury, Jamuna Kanta Sing, Dipak Kumar Basu, Mita
Nasipuri, Face recognition by generalized two dimensional FLD method
and multi-class support vector machines, Appl. Soft Comput. J. (2011)
[9] Keun-Chang Kwak; Pedrycz, W., &quot;Face Recognition Using an Enhanced
Independent Component Analysis Approach,&quot; Neural Networks, IEEE
Transactions on , vol.18, no.2, pp.530,541, March 2007
[10] Dagher, I.; Nachar, R., &quot;Face recognition using IPCA-ICA
algorithm,&quot; Pattern Analysis and Machine Intelligence, IEEE
Transactions on , vol.28, no.6, pp.996,1000, June 2006
[11] O. Deniz, M. Castrillon, M. Hernandez, &ldquo;Face recognition using
independent component analysis and support vector machines&rdquo;, Pattern
Recognition Letters, v.24 n.13, p.2153 &ndash; 2157, September 2003
[12] Dao-Qing Dai; Yuen, P.C., &quot;Face Recognition by Regularized
Discriminant Analysis,&quot; Systems, Man, and Cybernetics, Part B:
Cybernetics, IEEE Transactions on , vol.37, no.4, pp.1080,1085, Aug.
2007
[13] J. W. Lu , K. N. Plataniotis and A. N. Venetsanopoulos &quot;Regularization
studies of linear discriminant analysis in small sample size scenarios with
application to face recognition&quot;, Pattern Recognit. Lett., vol. 26, no.
2, pp.181 -191 2005.</p>