Abstract

Automatic target recognition in uncontrolled conditions is a difficult task because many parameters are involved. This study deals with the recognition of targets under limited out-of-plane rotations while maintaining invariance to ambient light illumination. Contrast invariance is achieved by using the recently developed locally adaptive contrast-invariant filter, a method that yields correlation peaks whose values are invariant under any linear transformation of intensity. To reduce the sensitivity to the orientation of the object we replace the reference in the nonlinear filter by a synthetic discriminant filter. The range used for out-of-plane rotations was 40 degrees with a depression angle of 20 degrees. We present results for unsegmented targets on complex backgrounds with the presence of false targets.

© 2003 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. F. M. Dickey, L. A. Romero, “Normalized correlation for pattern recognition,” Opt. Lett. 16, 1186–1188 (1991).
    [CrossRef] [PubMed]
  2. B. A. Kast, F. M. Dickey, “Normalization of correlators,” in Optical Information Processing Systems and Architectures III, B. Javidi, ed., Proc. Soc. Photo-Opt. Instrum. Eng.1564, 34–42 (1991).
  3. R. Kotynski, K. Chalasinska-Macukow, “Normalization of correlation filter based on the Hölder’s inequality,” in Optics in Computing ’98, P. H. Chavel, D. A. Miller, H. Thienpont, eds., Proc. Soc. Photo Opt. Instrum. Eng.3490, 195–198 (1998).
  4. R. Kotynski, K. Chalasinska-Macukow, “Multi-object intensity-invariant pattern recognition with an optimum processor for correlated noise,” in 18th Congress of the International Commission for Optics, A. J. Glass, J. W. Goodman, M. Chang, A. H. Guenther, eds., Proc. Soc. Photo-Opt. Instrum. Eng.3749, 316–317 (1999).
  5. H. H. Arsenault, C. Belisle, “Contrast-invariant pattern recognition using circular harmonic component,” Appl. Opt. 24, 2072–2075 (1985).
    [CrossRef]
  6. P. Garcia-Martinez, H. H. Arsenault, C. Ferreira, “Binary image decomposition for intensity-invariant optical nonlinear correlation,” in Optics in Computing 2000, R. A. Lessard, T. V. Galstian, eds., Proc. Soc. Photo-Opt. Instrum. Eng.4089, 433–438 (2000).
  7. H. H. Arsenault, D. Lefebvre, “Homomorphic cameo filter for pattern recognition that is invariant with change of illumination,” Opt. Lett. 25, 1567–1569 (2000).
    [CrossRef]
  8. K. Chalasinska-Macukow, F. Turon, M. J. Yzuel, J. Campos, “Contrast performance of pure phase correlation,” J. Opt. Soc. Am. A 24, 71–75 (1993).
  9. C. Minetti, F. Dubois, Jean-Claude Legros, “Reduction of the correlation sensitivity to the changes of the input illumination by post processing based on the correlation statistics,” Appl. Opt. 41, 3453–3460 (2002).
    [CrossRef] [PubMed]
  10. S. Zhang, M. A. Karim, “Illumination invariant pattern recognition with joint-transform-correlator-based morphological correlation,” Appl. Opt. 38, 7228–7237 (1999).
    [CrossRef]
  11. H. H. Arsenault, Y. Sheng, J. Bulabois, “Modified composite filter for pattern recognition in presence of noise with a non-zero mean,” Opt. Commun. 63, 15–20 (1987).
    [CrossRef]
  12. D. Lefebvre, H. H. Arsenault, P. Garcia-Martinez, C. Ferreira, “Recognition of unsegmented targets invariant under transformations of intensity,” Appl. Opt. 41, 6135–6142 (2002).
    [CrossRef] [PubMed]
  13. J. B. Burns, R. S. Weiss, E. M. Riseman, “View variation of point-set and line-segment features,” IEEE Transactions on Pattern Anal. Mach. Intell. PAMI-15, 51–68 (1993).
    [CrossRef]
  14. D. Casasent, W. Rozzi, D. Fetterly, “Correlation synthetic discriminant functions for object recognition and classification in high clutter,” Proc. SPIE 575, 126–136 (1985).
    [CrossRef]
  15. Y. Li, J. Rosen, “Three-dimensional pattern recognition with a single two-dimensional synthetic reference function,” Appl. Opt. 39, 1251–1259 (2000).
    [CrossRef]
  16. J. Rosen, “Three-dimensional electro-optical correlation,” J. Opt. Soc. Am. A 15, 430–436 (1998).
    [CrossRef]
  17. J. Rosen, “Three-dimensional joint transform correlator,” Appl. Opt. 37, 7538–7544 (1998).
    [CrossRef]
  18. B. Javidi, J. Li, Q. Tang, “Optical implementation of neural networks for face recognition by the use of nonlinear joint transform correlators,” Appl. Opt. 34, 3950–3962 (1995).
    [CrossRef] [PubMed]
  19. L. Neiberg, D. P. Casasent, “Feature space trajectory (FST) classifier neural network,” in Intelligent Robots and Computer Vision 13: Algorithms and Computer Vision, D. P. Casasent, ed., Proc. SPIE2353, 276–292 (1994).
    [CrossRef]
  20. S. Roy, H. H. Arsenault, “Shift, scale and pose invariant object recognition using wedge sampling and a feature space trajectory classifier,” J. Mod. Opt. 50, 285–297 (2003).
  21. S. Roy, H. H. Arsenault, D. Lefebvre, “Invariant object recognition under three-dimensional rotations and changes of scale,” Opt. Eng. 42, 813–821 (2003).
    [CrossRef]

2003 (2)

S. Roy, H. H. Arsenault, “Shift, scale and pose invariant object recognition using wedge sampling and a feature space trajectory classifier,” J. Mod. Opt. 50, 285–297 (2003).

S. Roy, H. H. Arsenault, D. Lefebvre, “Invariant object recognition under three-dimensional rotations and changes of scale,” Opt. Eng. 42, 813–821 (2003).
[CrossRef]

2002 (2)

2000 (2)

1999 (1)

1998 (2)

1995 (1)

1993 (2)

K. Chalasinska-Macukow, F. Turon, M. J. Yzuel, J. Campos, “Contrast performance of pure phase correlation,” J. Opt. Soc. Am. A 24, 71–75 (1993).

J. B. Burns, R. S. Weiss, E. M. Riseman, “View variation of point-set and line-segment features,” IEEE Transactions on Pattern Anal. Mach. Intell. PAMI-15, 51–68 (1993).
[CrossRef]

1991 (1)

1987 (1)

H. H. Arsenault, Y. Sheng, J. Bulabois, “Modified composite filter for pattern recognition in presence of noise with a non-zero mean,” Opt. Commun. 63, 15–20 (1987).
[CrossRef]

1985 (2)

D. Casasent, W. Rozzi, D. Fetterly, “Correlation synthetic discriminant functions for object recognition and classification in high clutter,” Proc. SPIE 575, 126–136 (1985).
[CrossRef]

H. H. Arsenault, C. Belisle, “Contrast-invariant pattern recognition using circular harmonic component,” Appl. Opt. 24, 2072–2075 (1985).
[CrossRef]

Arsenault, H. H.

S. Roy, H. H. Arsenault, D. Lefebvre, “Invariant object recognition under three-dimensional rotations and changes of scale,” Opt. Eng. 42, 813–821 (2003).
[CrossRef]

S. Roy, H. H. Arsenault, “Shift, scale and pose invariant object recognition using wedge sampling and a feature space trajectory classifier,” J. Mod. Opt. 50, 285–297 (2003).

D. Lefebvre, H. H. Arsenault, P. Garcia-Martinez, C. Ferreira, “Recognition of unsegmented targets invariant under transformations of intensity,” Appl. Opt. 41, 6135–6142 (2002).
[CrossRef] [PubMed]

H. H. Arsenault, D. Lefebvre, “Homomorphic cameo filter for pattern recognition that is invariant with change of illumination,” Opt. Lett. 25, 1567–1569 (2000).
[CrossRef]

H. H. Arsenault, Y. Sheng, J. Bulabois, “Modified composite filter for pattern recognition in presence of noise with a non-zero mean,” Opt. Commun. 63, 15–20 (1987).
[CrossRef]

H. H. Arsenault, C. Belisle, “Contrast-invariant pattern recognition using circular harmonic component,” Appl. Opt. 24, 2072–2075 (1985).
[CrossRef]

P. Garcia-Martinez, H. H. Arsenault, C. Ferreira, “Binary image decomposition for intensity-invariant optical nonlinear correlation,” in Optics in Computing 2000, R. A. Lessard, T. V. Galstian, eds., Proc. Soc. Photo-Opt. Instrum. Eng.4089, 433–438 (2000).

Belisle, C.

Bulabois, J.

H. H. Arsenault, Y. Sheng, J. Bulabois, “Modified composite filter for pattern recognition in presence of noise with a non-zero mean,” Opt. Commun. 63, 15–20 (1987).
[CrossRef]

Burns, J. B.

J. B. Burns, R. S. Weiss, E. M. Riseman, “View variation of point-set and line-segment features,” IEEE Transactions on Pattern Anal. Mach. Intell. PAMI-15, 51–68 (1993).
[CrossRef]

Campos, J.

K. Chalasinska-Macukow, F. Turon, M. J. Yzuel, J. Campos, “Contrast performance of pure phase correlation,” J. Opt. Soc. Am. A 24, 71–75 (1993).

Casasent, D.

D. Casasent, W. Rozzi, D. Fetterly, “Correlation synthetic discriminant functions for object recognition and classification in high clutter,” Proc. SPIE 575, 126–136 (1985).
[CrossRef]

Casasent, D. P.

L. Neiberg, D. P. Casasent, “Feature space trajectory (FST) classifier neural network,” in Intelligent Robots and Computer Vision 13: Algorithms and Computer Vision, D. P. Casasent, ed., Proc. SPIE2353, 276–292 (1994).
[CrossRef]

Chalasinska-Macukow, K.

K. Chalasinska-Macukow, F. Turon, M. J. Yzuel, J. Campos, “Contrast performance of pure phase correlation,” J. Opt. Soc. Am. A 24, 71–75 (1993).

R. Kotynski, K. Chalasinska-Macukow, “Multi-object intensity-invariant pattern recognition with an optimum processor for correlated noise,” in 18th Congress of the International Commission for Optics, A. J. Glass, J. W. Goodman, M. Chang, A. H. Guenther, eds., Proc. Soc. Photo-Opt. Instrum. Eng.3749, 316–317 (1999).

R. Kotynski, K. Chalasinska-Macukow, “Normalization of correlation filter based on the Hölder’s inequality,” in Optics in Computing ’98, P. H. Chavel, D. A. Miller, H. Thienpont, eds., Proc. Soc. Photo Opt. Instrum. Eng.3490, 195–198 (1998).

Dickey, F. M.

F. M. Dickey, L. A. Romero, “Normalized correlation for pattern recognition,” Opt. Lett. 16, 1186–1188 (1991).
[CrossRef] [PubMed]

B. A. Kast, F. M. Dickey, “Normalization of correlators,” in Optical Information Processing Systems and Architectures III, B. Javidi, ed., Proc. Soc. Photo-Opt. Instrum. Eng.1564, 34–42 (1991).

Dubois, F.

Ferreira, C.

D. Lefebvre, H. H. Arsenault, P. Garcia-Martinez, C. Ferreira, “Recognition of unsegmented targets invariant under transformations of intensity,” Appl. Opt. 41, 6135–6142 (2002).
[CrossRef] [PubMed]

P. Garcia-Martinez, H. H. Arsenault, C. Ferreira, “Binary image decomposition for intensity-invariant optical nonlinear correlation,” in Optics in Computing 2000, R. A. Lessard, T. V. Galstian, eds., Proc. Soc. Photo-Opt. Instrum. Eng.4089, 433–438 (2000).

Fetterly, D.

D. Casasent, W. Rozzi, D. Fetterly, “Correlation synthetic discriminant functions for object recognition and classification in high clutter,” Proc. SPIE 575, 126–136 (1985).
[CrossRef]

Garcia-Martinez, P.

D. Lefebvre, H. H. Arsenault, P. Garcia-Martinez, C. Ferreira, “Recognition of unsegmented targets invariant under transformations of intensity,” Appl. Opt. 41, 6135–6142 (2002).
[CrossRef] [PubMed]

P. Garcia-Martinez, H. H. Arsenault, C. Ferreira, “Binary image decomposition for intensity-invariant optical nonlinear correlation,” in Optics in Computing 2000, R. A. Lessard, T. V. Galstian, eds., Proc. Soc. Photo-Opt. Instrum. Eng.4089, 433–438 (2000).

Javidi, B.

Karim, M. A.

Kast, B. A.

B. A. Kast, F. M. Dickey, “Normalization of correlators,” in Optical Information Processing Systems and Architectures III, B. Javidi, ed., Proc. Soc. Photo-Opt. Instrum. Eng.1564, 34–42 (1991).

Kotynski, R.

R. Kotynski, K. Chalasinska-Macukow, “Multi-object intensity-invariant pattern recognition with an optimum processor for correlated noise,” in 18th Congress of the International Commission for Optics, A. J. Glass, J. W. Goodman, M. Chang, A. H. Guenther, eds., Proc. Soc. Photo-Opt. Instrum. Eng.3749, 316–317 (1999).

R. Kotynski, K. Chalasinska-Macukow, “Normalization of correlation filter based on the Hölder’s inequality,” in Optics in Computing ’98, P. H. Chavel, D. A. Miller, H. Thienpont, eds., Proc. Soc. Photo Opt. Instrum. Eng.3490, 195–198 (1998).

Lefebvre, D.

Legros, Jean-Claude

Li, J.

Li, Y.

Minetti, C.

Neiberg, L.

L. Neiberg, D. P. Casasent, “Feature space trajectory (FST) classifier neural network,” in Intelligent Robots and Computer Vision 13: Algorithms and Computer Vision, D. P. Casasent, ed., Proc. SPIE2353, 276–292 (1994).
[CrossRef]

Riseman, E. M.

J. B. Burns, R. S. Weiss, E. M. Riseman, “View variation of point-set and line-segment features,” IEEE Transactions on Pattern Anal. Mach. Intell. PAMI-15, 51–68 (1993).
[CrossRef]

Romero, L. A.

Rosen, J.

Roy, S.

S. Roy, H. H. Arsenault, D. Lefebvre, “Invariant object recognition under three-dimensional rotations and changes of scale,” Opt. Eng. 42, 813–821 (2003).
[CrossRef]

S. Roy, H. H. Arsenault, “Shift, scale and pose invariant object recognition using wedge sampling and a feature space trajectory classifier,” J. Mod. Opt. 50, 285–297 (2003).

Rozzi, W.

D. Casasent, W. Rozzi, D. Fetterly, “Correlation synthetic discriminant functions for object recognition and classification in high clutter,” Proc. SPIE 575, 126–136 (1985).
[CrossRef]

Sheng, Y.

H. H. Arsenault, Y. Sheng, J. Bulabois, “Modified composite filter for pattern recognition in presence of noise with a non-zero mean,” Opt. Commun. 63, 15–20 (1987).
[CrossRef]

Tang, Q.

Turon, F.

K. Chalasinska-Macukow, F. Turon, M. J. Yzuel, J. Campos, “Contrast performance of pure phase correlation,” J. Opt. Soc. Am. A 24, 71–75 (1993).

Weiss, R. S.

J. B. Burns, R. S. Weiss, E. M. Riseman, “View variation of point-set and line-segment features,” IEEE Transactions on Pattern Anal. Mach. Intell. PAMI-15, 51–68 (1993).
[CrossRef]

Yzuel, M. J.

K. Chalasinska-Macukow, F. Turon, M. J. Yzuel, J. Campos, “Contrast performance of pure phase correlation,” J. Opt. Soc. Am. A 24, 71–75 (1993).

Zhang, S.

Appl. Opt. (7)

IEEE Transactions on Pattern Anal. Mach. Intell. (1)

J. B. Burns, R. S. Weiss, E. M. Riseman, “View variation of point-set and line-segment features,” IEEE Transactions on Pattern Anal. Mach. Intell. PAMI-15, 51–68 (1993).
[CrossRef]

J. Mod. Opt. (1)

S. Roy, H. H. Arsenault, “Shift, scale and pose invariant object recognition using wedge sampling and a feature space trajectory classifier,” J. Mod. Opt. 50, 285–297 (2003).

J. Opt. Soc. Am. A (2)

J. Rosen, “Three-dimensional electro-optical correlation,” J. Opt. Soc. Am. A 15, 430–436 (1998).
[CrossRef]

K. Chalasinska-Macukow, F. Turon, M. J. Yzuel, J. Campos, “Contrast performance of pure phase correlation,” J. Opt. Soc. Am. A 24, 71–75 (1993).

Opt. Commun. (1)

H. H. Arsenault, Y. Sheng, J. Bulabois, “Modified composite filter for pattern recognition in presence of noise with a non-zero mean,” Opt. Commun. 63, 15–20 (1987).
[CrossRef]

Opt. Eng. (1)

S. Roy, H. H. Arsenault, D. Lefebvre, “Invariant object recognition under three-dimensional rotations and changes of scale,” Opt. Eng. 42, 813–821 (2003).
[CrossRef]

Opt. Lett. (2)

Proc. SPIE (1)

D. Casasent, W. Rozzi, D. Fetterly, “Correlation synthetic discriminant functions for object recognition and classification in high clutter,” Proc. SPIE 575, 126–136 (1985).
[CrossRef]

Other (5)

L. Neiberg, D. P. Casasent, “Feature space trajectory (FST) classifier neural network,” in Intelligent Robots and Computer Vision 13: Algorithms and Computer Vision, D. P. Casasent, ed., Proc. SPIE2353, 276–292 (1994).
[CrossRef]

B. A. Kast, F. M. Dickey, “Normalization of correlators,” in Optical Information Processing Systems and Architectures III, B. Javidi, ed., Proc. Soc. Photo-Opt. Instrum. Eng.1564, 34–42 (1991).

R. Kotynski, K. Chalasinska-Macukow, “Normalization of correlation filter based on the Hölder’s inequality,” in Optics in Computing ’98, P. H. Chavel, D. A. Miller, H. Thienpont, eds., Proc. Soc. Photo Opt. Instrum. Eng.3490, 195–198 (1998).

R. Kotynski, K. Chalasinska-Macukow, “Multi-object intensity-invariant pattern recognition with an optimum processor for correlated noise,” in 18th Congress of the International Commission for Optics, A. J. Glass, J. W. Goodman, M. Chang, A. H. Guenther, eds., Proc. Soc. Photo-Opt. Instrum. Eng.3749, 316–317 (1999).

P. Garcia-Martinez, H. H. Arsenault, C. Ferreira, “Binary image decomposition for intensity-invariant optical nonlinear correlation,” in Optics in Computing 2000, R. A. Lessard, T. V. Galstian, eds., Proc. Soc. Photo-Opt. Instrum. Eng.4089, 433–438 (2000).

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 (4)

Fig. 1
Fig. 1

Correlation peak as a function of the number of objects L and of the cross correlation k.

Fig. 2
Fig. 2

Training set used for the experiments.

Fig. 3
Fig. 3

Scene with 4 objects to identify and with 12 false targets.

Fig. 4
Fig. 4

The correlation plane obtained with our method.

Tables (1)

Tables Icon

Table 1 Correlation Peak Values for Each Class of Object

Equations (17)

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

Tx=afx+b,
sx*f0x2,
f0x=fx-μf,
sx2*xN-sx*x2N2,
1Nf0x*f0x×sx*f0x21/Nsx2*x-1/N2sx*x2.
hx=i=1L wif0ix f01x*f01x|x=0=1 f02x*f02x|x=0=1  f0Lx*f0Lx|x=0=1.
hx=i=1L wif0ix=i=1L wif0ix=0.
x=i=1L ix.
pi=1hx*hxx=0f0ix*f0ixx=0.
fˆ0ix=f0ixf0ix*f0ix|x=0.
pi=1hx*hxx=0.
p=hth-1.
p=1tQ-11-1,
p=1tQ-11-1=1tQ11tQ1-11tQ-11-1=1tQ111Q-111tQ1-1=1tQ111Q-1MQ1-1,
p=1tQ11tQ-1QM1-1=1tQ11tM1-1=1tQ1L2=Q.
p=L-1k+1L.
Q=1.000.750.580.470.460.751.000.770.590.550.580.771.000.810.660.470.590.811.000.840.460.550.660.841.00.

Metrics