Abstract

Pattern recognition with high discrimination can be achieved with a morphological correlator. A modification of this correlator is carried out by use of a binary slicing process instead of linear thresholding. Although the obtained correlation result is not identical to the conventional morphological correlation, it requires fewer calculations and provides even higher discrimination. Two optical experimental implementations of this modified morphological correlator as well as some experimental results are shown.

© 1999 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. A. VanderLugt, Optical Signal Processing (Wiley, New York, 1992).
  2. J. W. Goodman, Introduction to Fourier Optics, 2nd ed. (Mc-Graw Hill, New York, 1996), pp. 237–246.
  3. P. Maragos, “Morphological correlation and mean absolute error criteria,” in Proceedings of the IEEE ICASSP—International Conference on Acoustics, Speech, and Signal Processing (Institute of Electrical and Electronics Engineers, New York, 1989), pp. 1568–1571.
    [CrossRef]
  4. P. Maragos, “Pattern spectrum and multiscale shape representation,” IEEE Trans. Pattern Anal. Mach. Intell. 11, 701–716 (1989).
    [CrossRef]
  5. P. Garcia-Martinez, D. Mas, J. Garcia, C. Ferreira, “Nonlinear morphological correlation: optoelectronic implementation,” Appl. Opt. 37, 2112–2118 (1998).
    [CrossRef]
  6. G. J. Swanson, “Binary optics technology: the theory and design of multilevel diffractive optical elements,” (Lincoln Laboratory, MIT, Cambridge, Mass., 1989).
  7. E. Ochoa, J. P. Allenbach, D. W. Sweeney, “Optical median filtering using threshold decomposition,” Appl. Opt. 26, 252–260 (1987).
    [CrossRef] [PubMed]
  8. M. Deutch, J. Garcia, D. Mendlovic, “Multichannel single-output color pattern recognition using a joint transform correlator,” Appl. Opt. 35, 6976–6982 (1996).
    [CrossRef]

1998 (1)

1996 (1)

1989 (1)

P. Maragos, “Pattern spectrum and multiscale shape representation,” IEEE Trans. Pattern Anal. Mach. Intell. 11, 701–716 (1989).
[CrossRef]

1987 (1)

Allenbach, J. P.

Deutch, M.

Ferreira, C.

Garcia, J.

Garcia-Martinez, P.

Goodman, J. W.

J. W. Goodman, Introduction to Fourier Optics, 2nd ed. (Mc-Graw Hill, New York, 1996), pp. 237–246.

Maragos, P.

P. Maragos, “Pattern spectrum and multiscale shape representation,” IEEE Trans. Pattern Anal. Mach. Intell. 11, 701–716 (1989).
[CrossRef]

P. Maragos, “Morphological correlation and mean absolute error criteria,” in Proceedings of the IEEE ICASSP—International Conference on Acoustics, Speech, and Signal Processing (Institute of Electrical and Electronics Engineers, New York, 1989), pp. 1568–1571.
[CrossRef]

Mas, D.

Mendlovic, D.

Ochoa, E.

Swanson, G. J.

G. J. Swanson, “Binary optics technology: the theory and design of multilevel diffractive optical elements,” (Lincoln Laboratory, MIT, Cambridge, Mass., 1989).

Sweeney, D. W.

VanderLugt, A.

A. VanderLugt, Optical Signal Processing (Wiley, New York, 1992).

Appl. Opt. (3)

IEEE Trans. Pattern Anal. Mach. Intell. (1)

P. Maragos, “Pattern spectrum and multiscale shape representation,” IEEE Trans. Pattern Anal. Mach. Intell. 11, 701–716 (1989).
[CrossRef]

Other (4)

G. J. Swanson, “Binary optics technology: the theory and design of multilevel diffractive optical elements,” (Lincoln Laboratory, MIT, Cambridge, Mass., 1989).

A. VanderLugt, Optical Signal Processing (Wiley, New York, 1992).

J. W. Goodman, Introduction to Fourier Optics, 2nd ed. (Mc-Graw Hill, New York, 1996), pp. 237–246.

P. Maragos, “Morphological correlation and mean absolute error criteria,” in Proceedings of the IEEE ICASSP—International Conference on Acoustics, Speech, and Signal Processing (Institute of Electrical and Electronics Engineers, New York, 1989), pp. 1568–1571.
[CrossRef]

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 (10)

Fig. 1
Fig. 1

(a) The reference image. (b) The corresponding masks of the reference image.

Fig. 2
Fig. 2

(a) The image masks when the intensity level is 127. (b) The corresponding masks when a unit-value change has occurred.

Fig. 3
Fig. 3

JTC scheme: SLM, spatial light modulator; F. T. lens, Fourier transform lens; J. T. plane, joint transform plane.

Fig. 4
Fig. 4

Spatial-multiplexing MMC approach: (a) input-plane arrangement and (b) responses obtained at the output plane.

Fig. 5
Fig. 5

Objects used for the experiments: (a) reference object and (b) observed scene.

Fig. 6
Fig. 6

Look-up table used for the autocorrelation test (solid curve and the cross-correlation test (dashed curve) with the transformed object.

Fig. 7
Fig. 7

Linear correlation results obtained with the reference object [Fig. 5(a)] and the input scheme [Fig. 5(b)]: (a) captured output correlation plane and (b) profile of the line connecting the two peaks.

Fig. 8
Fig. 8

Same as Fig. 7 but for the CMC.

Fig. 9
Fig. 9

Same as Fig. 7 but for the MMC.

Fig. 10
Fig. 10

Simulation results for the MMC by use of the spatial-multiplexing approach.

Equations (9)

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

MSEm=kRfk+m-gk2=kRfk+m2+gk2-2fk+mgk,
γfgm=kR fk+mgk.
MAEm=kR |fk+m-gk|.
|f-g|=f+g-2 minf, g
MAEm=kRfk+m+gk-2 minfk+m, gk,
μgfm=kmingk+m, fk=q=1Q gqm * fqm,
gqm=1gmq0gm<q.
βgfm=q=1Q gqm * fqm,
C1=g4 * f1=C21*,C2=g1 * g4+f1 * f4=C20*,C3=f4 * g1=C19*,C4=g3 * f1+g4 * f2=C18*,C5=g1 * g3+g2 * g4+f1 * f3+f2 * f4=C17*,C6=f3 * g1+f4 * g2=C16*,C7=g2 * f1+g3 * f2+g4 * f3=C15*,C8=g1 * g2+g2 * g3+g3 * g4+f1 * f2+f2 * f3+f3 * f4=C14*,C9=f2 * g1+f3 * g2+f4 * g3=C13*,C10=g1 * f1+g2 * f2+g3 * f3+g4 * f4=C12*,C11=Addition of all autocorrelation terms.

Metrics