A real-time image recognition system that can perform complex signal detection under partially coherent illumination is presented. The unique features of the correlator are that the target under study may be a diffusely illuminated real object, and the real-time correlation signal is produced using a narrowband extended light source. The system employs an optical correlator used with a spatial light modulator (liquid crystal image transducer). Pattern recognition experiments using naturally illuminated real objects are provided, and the effects of partially coherent illumination on the spatial phase variations in the output wave front of the device are investigated. Such nonuniformities may arise from nonflatness and mismatch of various output layers in some liquid crystal image transducers which can degrade the performance of the correlator. The experimental results indicate that the distortion effects present in the output surface of the device can be reduced by relaxing the coherence requirements of the system. A quantitative analysis is presented to study the effects of partially coherent illumination on the real-time correlation signal.
[1]
Pierre Chavel,et al.
Noise and coherence in optical image processing. II. Noise fluctuations
,
1978
.
[2]
G. Montgomery.
Optical properties of a liquid-crystal image transducer at normal incidence: Mathematical analysis and application to the off-state
,
1980
.
[3]
G. M. Morris,et al.
Matched filtering using band-limited illumination.
,
1980,
Optics letters.
[4]
F.T.S. Yu.
A new technique of incoherent complex signal detection
,
1978
.
[5]
A D Gara,et al.
Real-time optical correlation of 3-D scenes.
,
1977,
Applied optics.
[6]
Pierre Chavel.
Optical noise and temporal coherence
,
1980
.
[7]
H. B. Brown,et al.
Application Of The Liquid Crystal Light Valve To Real-Time Optical Data Processing
,
1978
.
[8]
F T Yu,et al.
Coherence requirements for partially coherent optical processing.
,
1982,
Applied optics.
[9]
A. B. Vander Lugt,et al.
Signal detection by complex spatial filtering
,
1964,
IEEE Trans. Inf. Theory.