Abstract

Face recognition on mobile devices, such as personal digital assistants and cell phones, is a big challenge owing to the limited computational resources available to run verifications on the devices themselves. One approach is to transmit the captured face images by use of the cell-phone connection and to run the verification on a remote station. However, owing to limitations in communication bandwidth, it may be necessary to transmit a compressed version of the image. We propose using the image compression standard JPEG2000, which is a wavelet-based compression engine used to compress the face images to low bit rates suitable for transmission over low-bandwidth communication channels. At the receiver end, the face images are reconstructed with a JPEG2000 decoder and are fed into the verification engine. We explore how advanced correlation filters, such as the minimum average correlation energy filter [ Appl. Opt. 26, 3633 ( 1987)] and its variants, perform by using face images captured under different illumination conditions and encoded with different bit rates under the JPEG2000 wavelet-encoding standard. We evaluate the performance of these filters by using illumination variations from the Carnegie Mellon University’s Pose, Illumination, and Expression (PIE) face database. We also demonstrate the tolerance of these filters to noisy versions of images with illumination variations.

© 2005 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. M. Savvides, B. V. K. Vijaya Kumar, “Efficient design of advanced correlation filters for robust distortion-tolerant face recognition,” in Proceedings of IEEE Conference on Advanced Video and Signal Based Surveillance (Institute of Electrical and Electronics Engineers, New York, 2003), pp. 45–52.
    [CrossRef]
  2. M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Authentication-invariant cancellable biometric filters for illumination-tolerant face verification,” in Biometric Technology for Human Identification, A. K. Jain, N. K. Ratha, eds., Proc. SPIE5404, 156–163 (2004).
    [CrossRef]
  3. M. Savvides, B. V. K. Vijaya Kumar, “Quad phase minimum average correlation energy filters for reduced memory illumination tolerant face authentication,” in Lecture Notes in Computer Science, G. Goos, J. Hartmanis, J. van Leeuwen, eds. (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
    [CrossRef]
  4. D. J. Moore, “JPEG2000 for handheld applications,” in Thirty-Seventh Asilomar Conference on Signals, Systems, and Computers (Institute of Electrical and Electronics Engineers, New York, 2003), Vol. 2, 1999–2003.
  5. A. Mahalanobis, B. V. K. Vijaya Kumar, D. Casasent, “Minimum average correlation energy filters,” Appl. Opt. 26, 3633–3640 (1987).
    [CrossRef] [PubMed]
  6. P. Réfrégier, “Optimal trade-off filters for noise robustness, sharpness of the correlation peak, and Horner efficiency,” Opt. Lett. 16, 829–831 (1991).
    [CrossRef] [PubMed]
  7. T. Sim, S. Baker, M. Bsat, “The CMU pose, illumination, and expression (PIE) database of human faces,” (Robotics Institute, Carnegie Mellon University, Pittsburgh, Pa., 2001).
  8. M. Unser, T. Blu, “Mathematical properties of the JPEG2000 wavelet filters,” IEEE Trans. Image Process. 12, 1080–1090 (2003).
    [CrossRef]
  9. J. Zhu, “Image compression using wavelets and JPEG2000: a tutorial,” Electron. Commun. Eng. J. 14, 112–121 (2002).
    [CrossRef]
  10. T. E. Diego Santa-Cruz, J. Askelöf, M. Larsson, C. Christopoulos, “JPEG2000 still image coding versus other standards,” in Applications of Digital Image Processing XXIII, A. G. Tescher, ed., Proc. SPIE4115, 446–454, 2000.
    [CrossRef]
  11. G. K. Wallace, “The JPEG still picture compression standard,” IEEE Trans. Consum. Electron. 38,xviii–xxxiv (1992).
    [CrossRef]
  12. B. E. Usevitch, “A tutorial on modern lossy wavelet image compression: foundations of JPEG 2000,” IEEE Signal Process. Mag. 18, 22–35 (2001).
    [CrossRef]
  13. J. M. Saphiro, “Embedded image coding using zerotrees of wavelet coefficients” IEEE Trans. Signal Process. 41, 3445–3462 (1993).
    [CrossRef]
  14. M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Face verification using correlation filters,” in Proceedings of the Third IEEE Automatic Identification Advanced Technology Conference (Institute of Electrical and Electronics Engineers, New York, 2002).
  15. M. Savvides, K. Venkataramani, B. V. K. Vijaya Kumar, “Incremental updating of advanced correlation filter designs,” in Proceedings of the International Conference on Multimedia and Expo (Institute of Electrical and Electronics Engineers, New York, 2003), Vol 3, pp. 229–232.
  16. M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Robust, shift-invariance biometric identification from partial face images,” in Biometric Technology for Human Identifications, A. K. Jain, K. Ratha, eds., Proc. SPIE5404, 124–135 (2004).
    [CrossRef]
  17. B. V. K. Vijaya Kumar, “Minimum variance synthetic discriminant functions,” J. Opt. Soc. Am. A 3, 1579–1584 (1986).
    [CrossRef]
  18. M. Savvides, B. V. K. Vijaya Kumar, “Illumination normalization using logarithm transforms for face authentication,” in Lecture Notes in Computer Science (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
    [CrossRef]
  19. A. Jain, Fundamentals of Digital Image Processing, Prentice-Hall Information and System Science Series, T. Kailath, ed. (Prentice-Hall, Englewood Cliffs, N.J., 1988).

2003 (1)

M. Unser, T. Blu, “Mathematical properties of the JPEG2000 wavelet filters,” IEEE Trans. Image Process. 12, 1080–1090 (2003).
[CrossRef]

2002 (1)

J. Zhu, “Image compression using wavelets and JPEG2000: a tutorial,” Electron. Commun. Eng. J. 14, 112–121 (2002).
[CrossRef]

2001 (1)

B. E. Usevitch, “A tutorial on modern lossy wavelet image compression: foundations of JPEG 2000,” IEEE Signal Process. Mag. 18, 22–35 (2001).
[CrossRef]

1993 (1)

J. M. Saphiro, “Embedded image coding using zerotrees of wavelet coefficients” IEEE Trans. Signal Process. 41, 3445–3462 (1993).
[CrossRef]

1992 (1)

G. K. Wallace, “The JPEG still picture compression standard,” IEEE Trans. Consum. Electron. 38,xviii–xxxiv (1992).
[CrossRef]

1991 (1)

1987 (1)

1986 (1)

Askelöf, J.

T. E. Diego Santa-Cruz, J. Askelöf, M. Larsson, C. Christopoulos, “JPEG2000 still image coding versus other standards,” in Applications of Digital Image Processing XXIII, A. G. Tescher, ed., Proc. SPIE4115, 446–454, 2000.
[CrossRef]

Baker, S.

T. Sim, S. Baker, M. Bsat, “The CMU pose, illumination, and expression (PIE) database of human faces,” (Robotics Institute, Carnegie Mellon University, Pittsburgh, Pa., 2001).

Blu, T.

M. Unser, T. Blu, “Mathematical properties of the JPEG2000 wavelet filters,” IEEE Trans. Image Process. 12, 1080–1090 (2003).
[CrossRef]

Bsat, M.

T. Sim, S. Baker, M. Bsat, “The CMU pose, illumination, and expression (PIE) database of human faces,” (Robotics Institute, Carnegie Mellon University, Pittsburgh, Pa., 2001).

Casasent, D.

Christopoulos, C.

T. E. Diego Santa-Cruz, J. Askelöf, M. Larsson, C. Christopoulos, “JPEG2000 still image coding versus other standards,” in Applications of Digital Image Processing XXIII, A. G. Tescher, ed., Proc. SPIE4115, 446–454, 2000.
[CrossRef]

Diego Santa-Cruz, T. E.

T. E. Diego Santa-Cruz, J. Askelöf, M. Larsson, C. Christopoulos, “JPEG2000 still image coding versus other standards,” in Applications of Digital Image Processing XXIII, A. G. Tescher, ed., Proc. SPIE4115, 446–454, 2000.
[CrossRef]

Jain, A.

A. Jain, Fundamentals of Digital Image Processing, Prentice-Hall Information and System Science Series, T. Kailath, ed. (Prentice-Hall, Englewood Cliffs, N.J., 1988).

Khosla, P. K.

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Face verification using correlation filters,” in Proceedings of the Third IEEE Automatic Identification Advanced Technology Conference (Institute of Electrical and Electronics Engineers, New York, 2002).

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Robust, shift-invariance biometric identification from partial face images,” in Biometric Technology for Human Identifications, A. K. Jain, K. Ratha, eds., Proc. SPIE5404, 124–135 (2004).
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Authentication-invariant cancellable biometric filters for illumination-tolerant face verification,” in Biometric Technology for Human Identification, A. K. Jain, N. K. Ratha, eds., Proc. SPIE5404, 156–163 (2004).
[CrossRef]

Larsson, M.

T. E. Diego Santa-Cruz, J. Askelöf, M. Larsson, C. Christopoulos, “JPEG2000 still image coding versus other standards,” in Applications of Digital Image Processing XXIII, A. G. Tescher, ed., Proc. SPIE4115, 446–454, 2000.
[CrossRef]

Mahalanobis, A.

Moore, D. J.

D. J. Moore, “JPEG2000 for handheld applications,” in Thirty-Seventh Asilomar Conference on Signals, Systems, and Computers (Institute of Electrical and Electronics Engineers, New York, 2003), Vol. 2, 1999–2003.

Réfrégier, P.

Saphiro, J. M.

J. M. Saphiro, “Embedded image coding using zerotrees of wavelet coefficients” IEEE Trans. Signal Process. 41, 3445–3462 (1993).
[CrossRef]

Savvides, M.

M. Savvides, B. V. K. Vijaya Kumar, “Quad phase minimum average correlation energy filters for reduced memory illumination tolerant face authentication,” in Lecture Notes in Computer Science, G. Goos, J. Hartmanis, J. van Leeuwen, eds. (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Authentication-invariant cancellable biometric filters for illumination-tolerant face verification,” in Biometric Technology for Human Identification, A. K. Jain, N. K. Ratha, eds., Proc. SPIE5404, 156–163 (2004).
[CrossRef]

M. Savvides, K. Venkataramani, B. V. K. Vijaya Kumar, “Incremental updating of advanced correlation filter designs,” in Proceedings of the International Conference on Multimedia and Expo (Institute of Electrical and Electronics Engineers, New York, 2003), Vol 3, pp. 229–232.

M. Savvides, B. V. K. Vijaya Kumar, “Efficient design of advanced correlation filters for robust distortion-tolerant face recognition,” in Proceedings of IEEE Conference on Advanced Video and Signal Based Surveillance (Institute of Electrical and Electronics Engineers, New York, 2003), pp. 45–52.
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, “Illumination normalization using logarithm transforms for face authentication,” in Lecture Notes in Computer Science (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Robust, shift-invariance biometric identification from partial face images,” in Biometric Technology for Human Identifications, A. K. Jain, K. Ratha, eds., Proc. SPIE5404, 124–135 (2004).
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Face verification using correlation filters,” in Proceedings of the Third IEEE Automatic Identification Advanced Technology Conference (Institute of Electrical and Electronics Engineers, New York, 2002).

Sim, T.

T. Sim, S. Baker, M. Bsat, “The CMU pose, illumination, and expression (PIE) database of human faces,” (Robotics Institute, Carnegie Mellon University, Pittsburgh, Pa., 2001).

Unser, M.

M. Unser, T. Blu, “Mathematical properties of the JPEG2000 wavelet filters,” IEEE Trans. Image Process. 12, 1080–1090 (2003).
[CrossRef]

Usevitch, B. E.

B. E. Usevitch, “A tutorial on modern lossy wavelet image compression: foundations of JPEG 2000,” IEEE Signal Process. Mag. 18, 22–35 (2001).
[CrossRef]

Venkataramani, K.

M. Savvides, K. Venkataramani, B. V. K. Vijaya Kumar, “Incremental updating of advanced correlation filter designs,” in Proceedings of the International Conference on Multimedia and Expo (Institute of Electrical and Electronics Engineers, New York, 2003), Vol 3, pp. 229–232.

Vijaya Kumar, B. V. K.

A. Mahalanobis, B. V. K. Vijaya Kumar, D. Casasent, “Minimum average correlation energy filters,” Appl. Opt. 26, 3633–3640 (1987).
[CrossRef] [PubMed]

B. V. K. Vijaya Kumar, “Minimum variance synthetic discriminant functions,” J. Opt. Soc. Am. A 3, 1579–1584 (1986).
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Authentication-invariant cancellable biometric filters for illumination-tolerant face verification,” in Biometric Technology for Human Identification, A. K. Jain, N. K. Ratha, eds., Proc. SPIE5404, 156–163 (2004).
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, “Quad phase minimum average correlation energy filters for reduced memory illumination tolerant face authentication,” in Lecture Notes in Computer Science, G. Goos, J. Hartmanis, J. van Leeuwen, eds. (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
[CrossRef]

M. Savvides, K. Venkataramani, B. V. K. Vijaya Kumar, “Incremental updating of advanced correlation filter designs,” in Proceedings of the International Conference on Multimedia and Expo (Institute of Electrical and Electronics Engineers, New York, 2003), Vol 3, pp. 229–232.

M. Savvides, B. V. K. Vijaya Kumar, “Illumination normalization using logarithm transforms for face authentication,” in Lecture Notes in Computer Science (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, “Efficient design of advanced correlation filters for robust distortion-tolerant face recognition,” in Proceedings of IEEE Conference on Advanced Video and Signal Based Surveillance (Institute of Electrical and Electronics Engineers, New York, 2003), pp. 45–52.
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Face verification using correlation filters,” in Proceedings of the Third IEEE Automatic Identification Advanced Technology Conference (Institute of Electrical and Electronics Engineers, New York, 2002).

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Robust, shift-invariance biometric identification from partial face images,” in Biometric Technology for Human Identifications, A. K. Jain, K. Ratha, eds., Proc. SPIE5404, 124–135 (2004).
[CrossRef]

Wallace, G. K.

G. K. Wallace, “The JPEG still picture compression standard,” IEEE Trans. Consum. Electron. 38,xviii–xxxiv (1992).
[CrossRef]

Zhu, J.

J. Zhu, “Image compression using wavelets and JPEG2000: a tutorial,” Electron. Commun. Eng. J. 14, 112–121 (2002).
[CrossRef]

Appl. Opt. (1)

Electron. Commun. Eng. J. (1)

J. Zhu, “Image compression using wavelets and JPEG2000: a tutorial,” Electron. Commun. Eng. J. 14, 112–121 (2002).
[CrossRef]

IEEE Signal Process. Mag. (1)

B. E. Usevitch, “A tutorial on modern lossy wavelet image compression: foundations of JPEG 2000,” IEEE Signal Process. Mag. 18, 22–35 (2001).
[CrossRef]

IEEE Trans. Consum. Electron. (1)

G. K. Wallace, “The JPEG still picture compression standard,” IEEE Trans. Consum. Electron. 38,xviii–xxxiv (1992).
[CrossRef]

IEEE Trans. Image Process. (1)

M. Unser, T. Blu, “Mathematical properties of the JPEG2000 wavelet filters,” IEEE Trans. Image Process. 12, 1080–1090 (2003).
[CrossRef]

IEEE Trans. Signal Process. (1)

J. M. Saphiro, “Embedded image coding using zerotrees of wavelet coefficients” IEEE Trans. Signal Process. 41, 3445–3462 (1993).
[CrossRef]

J. Opt. Soc. Am. A (1)

Opt. Lett. (1)

Other (11)

T. Sim, S. Baker, M. Bsat, “The CMU pose, illumination, and expression (PIE) database of human faces,” (Robotics Institute, Carnegie Mellon University, Pittsburgh, Pa., 2001).

T. E. Diego Santa-Cruz, J. Askelöf, M. Larsson, C. Christopoulos, “JPEG2000 still image coding versus other standards,” in Applications of Digital Image Processing XXIII, A. G. Tescher, ed., Proc. SPIE4115, 446–454, 2000.
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Face verification using correlation filters,” in Proceedings of the Third IEEE Automatic Identification Advanced Technology Conference (Institute of Electrical and Electronics Engineers, New York, 2002).

M. Savvides, K. Venkataramani, B. V. K. Vijaya Kumar, “Incremental updating of advanced correlation filter designs,” in Proceedings of the International Conference on Multimedia and Expo (Institute of Electrical and Electronics Engineers, New York, 2003), Vol 3, pp. 229–232.

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Robust, shift-invariance biometric identification from partial face images,” in Biometric Technology for Human Identifications, A. K. Jain, K. Ratha, eds., Proc. SPIE5404, 124–135 (2004).
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, “Efficient design of advanced correlation filters for robust distortion-tolerant face recognition,” in Proceedings of IEEE Conference on Advanced Video and Signal Based Surveillance (Institute of Electrical and Electronics Engineers, New York, 2003), pp. 45–52.
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, P. K. Khosla, “Authentication-invariant cancellable biometric filters for illumination-tolerant face verification,” in Biometric Technology for Human Identification, A. K. Jain, N. K. Ratha, eds., Proc. SPIE5404, 156–163 (2004).
[CrossRef]

M. Savvides, B. V. K. Vijaya Kumar, “Quad phase minimum average correlation energy filters for reduced memory illumination tolerant face authentication,” in Lecture Notes in Computer Science, G. Goos, J. Hartmanis, J. van Leeuwen, eds. (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
[CrossRef]

D. J. Moore, “JPEG2000 for handheld applications,” in Thirty-Seventh Asilomar Conference on Signals, Systems, and Computers (Institute of Electrical and Electronics Engineers, New York, 2003), Vol. 2, 1999–2003.

M. Savvides, B. V. K. Vijaya Kumar, “Illumination normalization using logarithm transforms for face authentication,” in Lecture Notes in Computer Science (Springer-Verlag, Berlin, 2003), Vol. 2680, pp. 19–26.
[CrossRef]

A. Jain, Fundamentals of Digital Image Processing, Prentice-Hall Information and System Science Series, T. Kailath, ed. (Prentice-Hall, Englewood Cliffs, N.J., 1988).

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 (16)

Fig. 1
Fig. 1

Three-level wavelet decomposition of a 1-D signal. Coeff i,j(n) refers to the ith decomposition level and the jth frequency band (j = 0 is the low-frequency subband, and j = 1 refers to the high-frequency subband).

Fig. 2
Fig. 2

(a) One-level wavelet tree decomposition of a face image of person 2 from the CMU PIE database. (b) Corresponding two-level wavelet tree decomposition of the same image.

Fig. 3
Fig. 3

Correlation filter block diagram shows a filter designed on N images for class i. When a test image from class i is input into the system, then the correlation output yields a sharp peak.

Fig. 4
Fig. 4

Region used for estimating the PSR.

Fig. 5
Fig. 5

Twenty-one samples of face images from person 2 that were captured under variable illumination conditions.

Fig. 6
Fig. 6

(a) Training set 1 uses extreme-illumination images 3, 7, and 16. (b) Training set 2 uses only near-frontal-illumination images 6, 7, and 8.

Fig. 7
Fig. 7

Reconstructed face image (7) of person 2 in the PIE database, compressed at various bit rates.

Fig. 8
Fig. 8

Threshold for FRR = 0 (when FAR = 0).

Fig. 9
Fig. 9

ROC (FAR versus FRR) curve for person 2 with training set 1 (images 3, 7, and 16; 100% verification is achieved for 1.5 −0.8 bpp).

Fig. 10
Fig. 10

ROC (FAR versus FRR) curve for person 2 with training set 2 (images 6, 7, and 8).

Fig. 11
Fig. 11

(a) Original image with severe self-shadowing. (b) Logarithm-transformed image.

Fig. 12
Fig. 12

White Gaussian noise added to the original image.

Fig. 13
Fig. 13

Original 7-dB SNR image and its reconstructed image at various bit-rate compressions.

Fig. 14
Fig. 14

Original 10-dB SNR image and its reconstructed image at various bit-rate compressions.

Fig. 15
Fig. 15

Original 15-dB SNR image and its reconstructed image at various bit-rate compressions.

Fig. 16
Fig. 16

ROC plot (FAR versus FRR) of person 2, using test images with (a) SNR = 7 dB, OTSDF α = 0.999; (b) 10 dB, OTSDF α = 0.9999; (c) SNR = 15 dB, OTSDF α = 0.99999. Images were compressed at the various bit rates shown, with the OTSDF filter trained with training set 1.

Tables (6)

Tables Icon

Table 1 Average Verification Rate (at 0% FAR) Achieved by Use of Test Images Compressed to Various Bit Rates with MACE Filters Synthesized with the Two Training Schemes

Tables Icon

Table 2 Average Verification Rate (at 0% FAR) Achieved by Use of Test Images Compressed at Various Bit Rates with OTSDF Filters Synthesized with Different Amounts of Noise Tolerance

Tables Icon

Table 3 Average Verification Performance of 65 People (at 0% FAR) with Compressed Logarithm-Transformed Test Images at Various Bit Rates

Tables Icon

Table 4 Verification Rates (at 0% FAR) of Test Images with 7-dB Additive White Gaussian Noise Compressed at Various Bit Rates and Tested with the OTSDF Filter at Different Amounts of Noise Tolerance

Tables Icon

Table 5 Verification Rates (at 0% FAR) of Test Images with 10-dB Additive White Gaussian Noise Compressed at Various Bit Rates and Tested with the OTSDF Filter at Different Amounts of Noise Tolerance

Tables Icon

Table 6 Verification Rates (at 0% FAR) of Test Images with 15-dB Additive White Gaussian Noise Compressed at Various Bit Rates and Tested with the OTSDF Filter at Different Amounts of Noise Tolerance

Equations (7)

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

c ( x , y ) = 2 - D InvFT { [ 2 - D FT { f ( x , y ) } ] H * ( u , v ) } .
E i = x = 0 d - 1 y = 0 d - 1 c i ( x , y ) 2 = 1 d 2 u = 0 d - 1 v = 0 d - 1 C i ( u , v ) c 2 = 1 d 2 u = 0 d - 1 v = 0 d - 1 H ( u , v ) 2 X i ( u , v ) 2 = h + D i h ,
E average = 1 N i = 1 N E i = 1 N i = 1 N h + D i h = h + Dh ,
X + h = u ,
h = D - 1 X ( X + D - 1 X ) - 1 u .
PSR = peak - mean σ .
h = T - 1 X ( X + T - 1 X ) - 1 u ,

Metrics