Abstract

We demonstrate a novel technique for face recognition. Our approach relies on the performances of a strongly discriminating optical correlation method along with the robustness of the independent component analysis (ICA) model. Simulations were performed to illustrate how this algorithm can identify a face with images from the Pointing Head Pose Image Database. While maintaining algorithmic simplicity, this approach based on ICA representation significantly increases the true recognition rate compared to that obtained using our previously developed all-numerical ICA identity recognition method and another method based on optical correlation and a standard composite filter.

© 2011 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. M. Turk and A. Pentland, J. Cogn. Neurosci. 3, 71 (1991).
    [CrossRef]
  2. L. Shen and L. Bai, Pattern Anal. Applic. 9, 273 (2006).
    [CrossRef]
  3. I. T. Jolliffe, Principal Component Analysis (Springer, 1986).
  4. P. Comon, Signal Process. 36, 287 (1994).
    [CrossRef]
  5. M. S. Bartlett, IEEE Trans. Neural Netw. 13, 1450 (2002).
    [CrossRef]
  6. A. Alfalou and C. Brosseau are preparing a manuscript to be titled “A comparative study of ICA face recognition techniques that use VLC correlation and ICA-based approaches.”
  7. B. V. K. V. Kumar, Appl. Opt. 31, 4773 (1992).
    [CrossRef] [PubMed]
  8. A. Alfalou, G. Keryer, and J. L. de Bougrenet de la Tocnaye, Appl. Opt. 38, 6129 (1999).
    [CrossRef]
  9. V. H. Diaz-Ramirez, Opt. Lasers Eng. 48, 1153 (2010).
    [CrossRef]
  10. A. Ahad and S. Awwal, Appl. Opt. 49, B40 (2010).
    [CrossRef]
  11. A. Alsamman and M. S. Alam, Appl. Opt. 44, 688 (2005).
    [CrossRef] [PubMed]
  12. A. Alfalou and A. Mansour, J. Opt. A 9, 787 (2007).
    [CrossRef]
  13. A. Alfalou, M. Farhat, and A. Mansour, in Proceedings of the 3rd IEEE International Conference on the Information and Communication Technology: From Theory to Applications (ICTA, 2008), doi: 10.1109/ICTTA.2008.4530111, pp. 1–4.
  14. J. L. Horner and P. D. Gianino, Appl. Opt. 23, 812 (1984).
    [CrossRef] [PubMed]
  15. J. L. Horner, Appl. Opt. 31, 165 (1992).
    [CrossRef] [PubMed]
  16. N. Gourier, D. Hall, and J. L. Crowley, in Proceedings of Pointing 2004, ICPR, International Workshop on Visual Observation of Deictic Gestures (2004), p. 1, http://www-prima.inrialpes.fr/Pointing04/data-face.html.
  17. J. P. Egan, Signal Detection Theory and ROC Analysis, Series in Cognition and Perception (Academic, 1975).

2010 (2)

V. H. Diaz-Ramirez, Opt. Lasers Eng. 48, 1153 (2010).
[CrossRef]

A. Ahad and S. Awwal, Appl. Opt. 49, B40 (2010).
[CrossRef]

2007 (1)

A. Alfalou and A. Mansour, J. Opt. A 9, 787 (2007).
[CrossRef]

2006 (1)

L. Shen and L. Bai, Pattern Anal. Applic. 9, 273 (2006).
[CrossRef]

2005 (1)

2002 (1)

M. S. Bartlett, IEEE Trans. Neural Netw. 13, 1450 (2002).
[CrossRef]

1999 (1)

1994 (1)

P. Comon, Signal Process. 36, 287 (1994).
[CrossRef]

1992 (2)

1991 (1)

M. Turk and A. Pentland, J. Cogn. Neurosci. 3, 71 (1991).
[CrossRef]

1984 (1)

Ahad, A.

Alam, M. S.

Alfalou, A.

A. Alfalou and A. Mansour, J. Opt. A 9, 787 (2007).
[CrossRef]

A. Alfalou, G. Keryer, and J. L. de Bougrenet de la Tocnaye, Appl. Opt. 38, 6129 (1999).
[CrossRef]

A. Alfalou and C. Brosseau are preparing a manuscript to be titled “A comparative study of ICA face recognition techniques that use VLC correlation and ICA-based approaches.”

A. Alfalou, M. Farhat, and A. Mansour, in Proceedings of the 3rd IEEE International Conference on the Information and Communication Technology: From Theory to Applications (ICTA, 2008), doi: 10.1109/ICTTA.2008.4530111, pp. 1–4.

Alsamman, A.

Awwal, S.

Bai, L.

L. Shen and L. Bai, Pattern Anal. Applic. 9, 273 (2006).
[CrossRef]

Bartlett, M. S.

M. S. Bartlett, IEEE Trans. Neural Netw. 13, 1450 (2002).
[CrossRef]

Brosseau, C.

A. Alfalou and C. Brosseau are preparing a manuscript to be titled “A comparative study of ICA face recognition techniques that use VLC correlation and ICA-based approaches.”

Comon, P.

P. Comon, Signal Process. 36, 287 (1994).
[CrossRef]

Crowley, J. L.

N. Gourier, D. Hall, and J. L. Crowley, in Proceedings of Pointing 2004, ICPR, International Workshop on Visual Observation of Deictic Gestures (2004), p. 1, http://www-prima.inrialpes.fr/Pointing04/data-face.html.

de Bougrenet de la Tocnaye, J. L.

Diaz-Ramirez, V. H.

V. H. Diaz-Ramirez, Opt. Lasers Eng. 48, 1153 (2010).
[CrossRef]

Egan, J. P.

J. P. Egan, Signal Detection Theory and ROC Analysis, Series in Cognition and Perception (Academic, 1975).

Farhat, M.

A. Alfalou, M. Farhat, and A. Mansour, in Proceedings of the 3rd IEEE International Conference on the Information and Communication Technology: From Theory to Applications (ICTA, 2008), doi: 10.1109/ICTTA.2008.4530111, pp. 1–4.

Gianino, P. D.

Gourier, N.

N. Gourier, D. Hall, and J. L. Crowley, in Proceedings of Pointing 2004, ICPR, International Workshop on Visual Observation of Deictic Gestures (2004), p. 1, http://www-prima.inrialpes.fr/Pointing04/data-face.html.

Hall, D.

N. Gourier, D. Hall, and J. L. Crowley, in Proceedings of Pointing 2004, ICPR, International Workshop on Visual Observation of Deictic Gestures (2004), p. 1, http://www-prima.inrialpes.fr/Pointing04/data-face.html.

Horner, J. L.

Jolliffe, I. T.

I. T. Jolliffe, Principal Component Analysis (Springer, 1986).

Keryer, G.

Kumar, B. V. K. V.

Mansour, A.

A. Alfalou and A. Mansour, J. Opt. A 9, 787 (2007).
[CrossRef]

A. Alfalou, M. Farhat, and A. Mansour, in Proceedings of the 3rd IEEE International Conference on the Information and Communication Technology: From Theory to Applications (ICTA, 2008), doi: 10.1109/ICTTA.2008.4530111, pp. 1–4.

Pentland, A.

M. Turk and A. Pentland, J. Cogn. Neurosci. 3, 71 (1991).
[CrossRef]

Shen, L.

L. Shen and L. Bai, Pattern Anal. Applic. 9, 273 (2006).
[CrossRef]

Turk, M.

M. Turk and A. Pentland, J. Cogn. Neurosci. 3, 71 (1991).
[CrossRef]

Appl. Opt. (6)

IEEE Trans. Neural Netw. (1)

M. S. Bartlett, IEEE Trans. Neural Netw. 13, 1450 (2002).
[CrossRef]

J. Cogn. Neurosci. (1)

M. Turk and A. Pentland, J. Cogn. Neurosci. 3, 71 (1991).
[CrossRef]

J. Opt. A (1)

A. Alfalou and A. Mansour, J. Opt. A 9, 787 (2007).
[CrossRef]

Opt. Lasers Eng. (1)

V. H. Diaz-Ramirez, Opt. Lasers Eng. 48, 1153 (2010).
[CrossRef]

Pattern Anal. Applic. (1)

L. Shen and L. Bai, Pattern Anal. Applic. 9, 273 (2006).
[CrossRef]

Signal Process. (1)

P. Comon, Signal Process. 36, 287 (1994).
[CrossRef]

Other (5)

I. T. Jolliffe, Principal Component Analysis (Springer, 1986).

A. Alfalou and C. Brosseau are preparing a manuscript to be titled “A comparative study of ICA face recognition techniques that use VLC correlation and ICA-based approaches.”

N. Gourier, D. Hall, and J. L. Crowley, in Proceedings of Pointing 2004, ICPR, International Workshop on Visual Observation of Deictic Gestures (2004), p. 1, http://www-prima.inrialpes.fr/Pointing04/data-face.html.

J. P. Egan, Signal Detection Theory and ROC Analysis, Series in Cognition and Perception (Academic, 1975).

A. Alfalou, M. Farhat, and A. Mansour, in Proceedings of the 3rd IEEE International Conference on the Information and Communication Technology: From Theory to Applications (ICTA, 2008), doi: 10.1109/ICTTA.2008.4530111, pp. 1–4.

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (3)

Fig. 1
Fig. 1

Schematics of the algorithm: (a) definition of the independent component learning base, (b) definition of the PCEs matrix, and (c) recognition procedure.

Fig. 2
Fig. 2

Sketch of the optical setup implementing the face recognition method using ICA.

Fig. 3
Fig. 3

Simulation results. (a) The curve in blue represents the identification results obtained with different facial positions of subject X, the curve in green corresponds to face Y ( 32 ) of subject Y. For both curves, three independent components were used. (b) Test base representing several positions of face X (the three faces of X used to compute the independent components are framed in red). (c) ROC plots using three independent components.

Equations (4)

Equations on this page are rendered with MathJax. Learn more.

S = a 1 C 1 + a 2 C 2 + + a n C n = i = 1 n a i C i ,
PCE i j = f ( V i pof j ) ,
PCE j = f ( V ? pof j ) = f ( ( b 1 C 1 + + b n C n ) C j ) = b j + N noise ,
ε i = | PCE 1 PCE i 1 | + | PCE 2 PCE i 2 | + + | PCE n PCE i n | .

Metrics