Abstract

A joint transform correlator has been constructed by a spatial light modulator that uses the electroabsorption effect of GaAs crystal and operates at a high frame rate a TV camera with logarithmic response and a personal computer. In the system, logarithmic values of joint power spectra generated in an optical system were electrically digitized and inverse-Fourier transformed. The system has accomplished the operation of correlation with a throughput time smaller than 10 ms per an input image.

© 2003 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. Y. Osugi, H. Mizukawa, T. Minemoto, “Quantization and truncation conditions of Fourier power spectrum for good performance in binary subtracted joint transform correlator,” Opt. Rev. 3, 161–170 (1996).
    [CrossRef]
  2. Y. Bitou, H. Ohta, T. Minemoto, “High-speed and high-contrast spatial light modulator that uses electroabsorption in a GaAs single crystal,” Appl. Opt. 37, 1377–1385 (1998).
    [CrossRef]
  3. Y. Bitou, T. Minemoto, “High-contrast spatial light modulator by use of the electroabsorption and electro-optic effects in a GaAs single crystal,” Appl. Opt. 37, 4347–4356 (1998).
    [CrossRef]
  4. Y. Bitou, Y. Osugi, T. Minemoto, “High-speed and high-contrast incoherent-to-coherent converter that uses a GaAs single crystal,” Opt. Eng. 38, 710–716 (1999).
    [CrossRef]
  5. Y. Iwamoto, T. Minemoto, “Comparisons of performance under brightness changes of target scenes between nonlinear joint transform correlators that use input images with pure-amplitude, pure-phase, and complex data,” Opt. Eng. 41, 41–49 (2002).
    [CrossRef]
  6. The TV camera system was made by Institute for Microelectronics Stuttgart, Germany.
  7. L. P. Yaroslavsky, “Nonlinear optical correlators with improved discrimination capability for object location and recognition,” in Optical Pattern Recognition, F. T. S. Yu, ed. (Cambridge University, Cambridge, U.K., 1998), Ch. 6, pp. 159–164.
  8. T. Yao, T. Minemoto, “Pattern recognition by quantization of modulated function to complexes of a quadrupole in matched filtering,” Opt. Eng. (to be published).

2002 (1)

Y. Iwamoto, T. Minemoto, “Comparisons of performance under brightness changes of target scenes between nonlinear joint transform correlators that use input images with pure-amplitude, pure-phase, and complex data,” Opt. Eng. 41, 41–49 (2002).
[CrossRef]

1999 (1)

Y. Bitou, Y. Osugi, T. Minemoto, “High-speed and high-contrast incoherent-to-coherent converter that uses a GaAs single crystal,” Opt. Eng. 38, 710–716 (1999).
[CrossRef]

1998 (2)

1996 (1)

Y. Osugi, H. Mizukawa, T. Minemoto, “Quantization and truncation conditions of Fourier power spectrum for good performance in binary subtracted joint transform correlator,” Opt. Rev. 3, 161–170 (1996).
[CrossRef]

Bitou, Y.

Iwamoto, Y.

Y. Iwamoto, T. Minemoto, “Comparisons of performance under brightness changes of target scenes between nonlinear joint transform correlators that use input images with pure-amplitude, pure-phase, and complex data,” Opt. Eng. 41, 41–49 (2002).
[CrossRef]

Minemoto, T.

Y. Iwamoto, T. Minemoto, “Comparisons of performance under brightness changes of target scenes between nonlinear joint transform correlators that use input images with pure-amplitude, pure-phase, and complex data,” Opt. Eng. 41, 41–49 (2002).
[CrossRef]

Y. Bitou, Y. Osugi, T. Minemoto, “High-speed and high-contrast incoherent-to-coherent converter that uses a GaAs single crystal,” Opt. Eng. 38, 710–716 (1999).
[CrossRef]

Y. Bitou, H. Ohta, T. Minemoto, “High-speed and high-contrast spatial light modulator that uses electroabsorption in a GaAs single crystal,” Appl. Opt. 37, 1377–1385 (1998).
[CrossRef]

Y. Bitou, T. Minemoto, “High-contrast spatial light modulator by use of the electroabsorption and electro-optic effects in a GaAs single crystal,” Appl. Opt. 37, 4347–4356 (1998).
[CrossRef]

Y. Osugi, H. Mizukawa, T. Minemoto, “Quantization and truncation conditions of Fourier power spectrum for good performance in binary subtracted joint transform correlator,” Opt. Rev. 3, 161–170 (1996).
[CrossRef]

T. Yao, T. Minemoto, “Pattern recognition by quantization of modulated function to complexes of a quadrupole in matched filtering,” Opt. Eng. (to be published).

Mizukawa, H.

Y. Osugi, H. Mizukawa, T. Minemoto, “Quantization and truncation conditions of Fourier power spectrum for good performance in binary subtracted joint transform correlator,” Opt. Rev. 3, 161–170 (1996).
[CrossRef]

Ohta, H.

Osugi, Y.

Y. Bitou, Y. Osugi, T. Minemoto, “High-speed and high-contrast incoherent-to-coherent converter that uses a GaAs single crystal,” Opt. Eng. 38, 710–716 (1999).
[CrossRef]

Y. Osugi, H. Mizukawa, T. Minemoto, “Quantization and truncation conditions of Fourier power spectrum for good performance in binary subtracted joint transform correlator,” Opt. Rev. 3, 161–170 (1996).
[CrossRef]

Yao, T.

T. Yao, T. Minemoto, “Pattern recognition by quantization of modulated function to complexes of a quadrupole in matched filtering,” Opt. Eng. (to be published).

Yaroslavsky, L. P.

L. P. Yaroslavsky, “Nonlinear optical correlators with improved discrimination capability for object location and recognition,” in Optical Pattern Recognition, F. T. S. Yu, ed. (Cambridge University, Cambridge, U.K., 1998), Ch. 6, pp. 159–164.

Appl. Opt. (2)

Opt. Eng. (2)

Y. Bitou, Y. Osugi, T. Minemoto, “High-speed and high-contrast incoherent-to-coherent converter that uses a GaAs single crystal,” Opt. Eng. 38, 710–716 (1999).
[CrossRef]

Y. Iwamoto, T. Minemoto, “Comparisons of performance under brightness changes of target scenes between nonlinear joint transform correlators that use input images with pure-amplitude, pure-phase, and complex data,” Opt. Eng. 41, 41–49 (2002).
[CrossRef]

Opt. Rev. (1)

Y. Osugi, H. Mizukawa, T. Minemoto, “Quantization and truncation conditions of Fourier power spectrum for good performance in binary subtracted joint transform correlator,” Opt. Rev. 3, 161–170 (1996).
[CrossRef]

Other (3)

The TV camera system was made by Institute for Microelectronics Stuttgart, Germany.

L. P. Yaroslavsky, “Nonlinear optical correlators with improved discrimination capability for object location and recognition,” in Optical Pattern Recognition, F. T. S. Yu, ed. (Cambridge University, Cambridge, U.K., 1998), Ch. 6, pp. 159–164.

T. Yao, T. Minemoto, “Pattern recognition by quantization of modulated function to complexes of a quadrupole in matched filtering,” Opt. Eng. (to be published).

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 (13)

Fig. 1
Fig. 1

(a) Facial images used in computer simulations and an experiment. (a) The output signals |o(x′, y′)|2 (b) and |l(x′, y′)|2 (c) obtained by computer simulations in the classical JTC and LJTC, respectively. In (a) the upper and the lower images were considered as the object and the reference images, respectively. Only halves of three-dimensional displays for the signals in the output planes are shown in (b) and (c). The sizes of the input and output planes in the simulations were 128 × 128 pixels.

Fig. 2
Fig. 2

Hybrid system of LJTC. Device: GaAs-SLM, DM: dichroic mirror, HF: half mirror, L: lens, and LD: laser diode.

Fig. 3
Fig. 3

Time schedules for operations of the elements in the hybrid system. T: the period of ac high voltage, τe: illuminating duration of erasing light, τw: effective write-in duration, τr: illuminating duration of readout light, t s: timing of trigger pulse for JPS image acquisition in HDRC camera.

Fig. 4
Fig. 4

Output signal experimentally obtained from the hybrid system for the input image shown in Fig. 1(a) under the conditions: V ac = 3.0 kV, f = 100 Hz, τe = 4.0 ms, τw = 0.4 ms, and τr = 2.0 ms.

Fig. 5
Fig. 5

Photograph of input image (a) and an illustration of correlation signals (b). In the input image, the circle pattern at the bottom was fixed and considered as the reference pattern. The other four patterns were attached on a rotating disk and considered as the object images. Three patterns of object images (a circle, a triangle, and an elliptic patterns) were rotating around the center of the rotating disk, and another circle pattern was attached at the center.

Fig. 6
Fig. 6

Change of the images recorded in GaAs-SLM during the erasing process, when the hybrid system was operating with V ac = 3.0 kV, f = 30 Hz, and no erasing light: (a), (b), and (c) capture the readout image just before, at 2 ms after starting, and at 5 ms after starting the erasing process, respectively. The left column shows the readout images and the right one the corresponding output signals.

Fig. 7
Fig. 7

Change of images recorded in GaAs-SLM during the erasing process, when the hybrid system was operating with V ac = 2.68 kV, f = 100 Hz, and no erasing light: (a), (b), and (c) capture the readout image just before, at 1.0 ms after starting, and at 2.5 ms after starting the erasing process, respectively. The left column shows the readout images and the right one the corresponding output signals.

Fig. 8
Fig. 8

Change of images recorded in GaAs-SLM during the erasing process, when the hybrid system was operating with V ac = 2.68 kV, f = 100 Hz, and τe = 4.0 ms: (a), (b), and (c) capture the readout image just before, at 1.2 ms after starting, and at 2.5 ms after starting the erasing process, respectively. The left column shows the readout images and the right one the corresponding output signals.

Fig. 9
Fig. 9

Correlation signals obtained from the hybrid system operating under the following conditions: V ac = 3.0 kV, f = 30 Hz, τw = 1.2 ms, τr = 2.8 ms, no erasing light, and the rotation of the disk = 35.6 rpm. (a), (b), and (c) shows the experimental results at the frame numbers 46, 49, and 53, respectively.

Fig. 10
Fig. 10

Correlation signals obtained from the hybrid system operating under the following conditions: V ac = 2.68 kV, f = 100 Hz, τe = 4.0 ms, τw = 0.5 ms, τr = 0.9 ms, and the rotation of the disk = 35.6 rpm. (a), (b), and (c) show the experimental results at the frame numbers 155, 175, and 207, respectively.

Fig. 11
Fig. 11

Correlation signals obtained from the hybrid system operating under the following conditions: V ac = 3.0 kV, f = 30 Hz, τw = 1.2 ms, and τr = 2.8 ms, no erasing light, and the rotation of the disk = 60 rpm. (a), (b), and (c) show the experimental results at the frame numbers 12, 14, and 16, respectively.

Fig. 12
Fig. 12

Correlation signals obtained from the hybrid system operating under the following conditions: V ac = 2.68 kV, f = 100 Hz, τe = 4.0 ms, τw = 0.5 ms, and τr = 0.9 ms, and the rotation of the disk = 60 rpm. (a), (b), and (c) show the experimental results at the frame numbers 36, 46, and 53, respectively.

Fig. 13
Fig. 13

Trajectories of the autocorrelation signals A, B, and C, when the rotation of the circular disk was 60 rpm. An X with dots at the compass points and a cross express the positions of autocorrelation signals A and B, respectively, obtained by the hybrid system operating with V ac = 3.0 kV and f = 30 Hz, and the open circle and open square mark locations with V ac = 2.68 kV and f = 100 Hz. The open triangle expresses the autocorrelation signal C.

Tables (1)

Tables Icon

Table 1 Performance of the Logarithmic Joint Transform Correlators

Equations (9)

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

fx, y=rx-xr, y-yr+tx-xt, y-yt,
|Fu, v|2=|Ru, vexp-juxr+yvr+Tu, vexp-juxt+vyt|2 =|Ru, v|2+|Tu, v|2+Ru, vTu, v*exp-juxr-xt+vyr-yt+Ru, v*Tu, v×expjuxr-xt+vyr-yt,
ox, y=FT-1|Fu, v|2=rx, yrx, y+tx, ytx, y+rx, ytx, yδx-xr-xt,y-yr-yt+rx, ytx, yδx+xr-xt, y+yr-yt,
Lu, v=log|Fu, v|2=log|Ru, v|2+|Tu, v|2+log1+Mu, v,
Mu, v=Ru, vTu, v*exp-juxr-xt+vyr-yt+Ru, v*Tu, v×expjuxr-xt+vyr-yt/|Ru, v|2+|Tu, v|2
|Mu, v|1.
Lu, v=log|Ru, v|2+|Tu, v|2+Mu, v-12Mu, v2+13Mu, v3++-1kMu, vk/k+.
lx, y=FT-1Lu, v=FT-1log|Ru, v|2+|Tu, v|2+FT-1Mu, v-FT-112Mu, v2+FT-113Mu, v3++FT-1-1kMu, vkk+.
|Fu, v|2/|Ru, v|2+|Tu, v|2=1+Mu, v =1+{Ru, vTu, v*exp-juxr-xt+vyr-yt+Ru, v*Tu, vexpjuxr-xt+vyr-yt}/|Ru, v|2+|Tu, v|2.

Metrics