The USC Andrew and Erna Viterbi School of Engineering USC Signal and Image Processing Institute USC Ming Hsieh Department of Electrical and Computer Engineering University of Southern California

Technical Report USC-IPI-930

“Characterization of Liquid Crystal Light Valves and Their Applications to Real-Time Nonlinear Optical Processing”

by Jerry Dean Michaelson

October 1979

The use of optical processors to perform linear operations such as Fourier transforming, convolution, and correlation on large bandwidth signals and images has been of great interest and has been investigated for some time. Extension of the processing capability to include nonlinear operations has been the subject of more recent interest and investigation. In this dissertation a method is presented to realize real-time nonlinear operations on two dimensional images through the use of multiple liquid crystal light valves.

A review is given of the theory associated with the light valves. Consideration is given both to electrically controlled cells and photoactivated devices. Special characteristics of the light valves pertinent to the nonlinear processing system are considered in detail. Actual light valve measurements are compared to theoretical predictions to verify the validity of the theoretical device models. A hardware implementation of the nonlinear processing system is presented using multiple light valves. Experimental system results are given which include level slice, logarithm, exponentiation, and A/D conversion operations. Consideration is given to both one- and two-dimensional input signals. The limitations of the light valves as they apply to the processing system are discussed, and methods of optimizing the system performance are considered. The methods include optical feedback as a means of improving the response characteristics of the light valves.

To download the report in PDF format click here: USC-IPI-930.pdf (4.4Mb)