A review of coherent optical correlator architectures, techniques and pattern recognition applications is presented.

Visual display systems for training simulators have traditionally used large model boards for scene image storage. The inherent physical nature of model boards makes them an inefficient method of image storage. In addition, other physical problems severely limit the range of training scenarios.

Biological specimens are recognized with a matched spatial filter (MSF) made from their simulated pattern. The partial correlations between the simulated pattern and the various sized specimens show that the simplified filter is considerably less sensitive to the size variation while it still retains a high degree of pattern discrimination against other species.

This paper reports recently obtained experimental results in the implementation of multiplexed holographic representations of space-variant optical systems. Phase-coded reference beams are used to suppress crosstalk between the multiplexed holograms. Results obtained using both ground glass and binary amplitude-coded diffusers for encoding the reference beams used in recording the multiplexed holograms are presented. A holographic representation of a particular 2-D space-variant optical processor is compared with the actual system's output.

Results are presented of an initial program to determine the feasibility of the use of a coherent optical matched spatial filter correlator for location of all pages in a microfilm database on which a given key word occurs. A modified frequency plane correlator system is used to quantify the correlation degradation effects of various differences (font, point size, etc.) expected between text inputs. The use of weighted matched spatial filter synthesis to decrease these correlation degradation effects is also analyzed. Quantitative experimental data are also included.

The techniques used for storing matched filters in image processing units of the optical matched filter image correlator (OMFIC*) type determine to a large extent the capacity and discrimination of the complete system. In this progress report of an ongoing investigation sponsored by the U.S. Air Force Office of Scientific Research (AFOSR), techniques and materials for storing matched filters are reported upon. Results are related by performance in the areas of diffraction efficiency, and signal to clutter measurements with selected imaging of targets in terrain scenes. Coherent and noncoherent addition of images for overlayed matched filter recording was investigated and is reported upon. Dichromated gelatins are used as matched filter recording media and the results compared with silver salt photographic materials. A formula is derived for MF capacity and the analysis extended to the case where the medium sets the resolution for MF recording.

The image quality attainable in producing holographic components depends to a great extend on the optical quality of the holographic recording medium and eventually its supporting substrate. Departure from the assumed ideal optical homogeneity and perfect geometry leads to a loss of image resolution and diffraction efficiency. The problem is analysed by recourse to a wave representation in terms of spatial frequency spectrum. The wedge phase error and the sinuosoidal perturbation are studied. The obtained results permit to derive a tolerance range for the substrate quality. Experimental interferometric methods are presented to the test of holographic plates quality.

A method of inputting microscopic phase objects into a coherent optical pattern recognition system is described. The method consists of directing a laser beam into an interference contrast microscope to produce an enlarged, coherent image of the specimen. The system has the advantage of presenting high contrast, dark field images to the correlator without the need of additional pre-processing for image enhancement. Applications to diatom recognition are shown.

The objective of this paper is to provide an overview of the state of the art of optical storage media. General specification and -performance data are presented for a variety of relevant direct digital and holographic storage media. Observed trends are noted, and suggestions for development emphasis in the 1980's are made. An overall conclusion is that the performance of storage media has improved significantly over the past five years. With appropriate systems engineering, the storage medium no longer need slow or prevent the future development of competitive optical information handling systems.

The Microchannel spatial light modulator is an optically-addressed spatial light modulator that is being developed for low-light-level, real-time, optical signal processing. In addition to its framed or continuous real-time mode of operation, it can also be used for short-time (hours), low resolution (less than 20 cycles/mm) optical information storage. In this Paper we describe the structure of the device, its principles of operation, some of its preliminary characteristics, and the first attempts at optical information storage.

Dichromated gelatin is considered to be one of the best holographic storage media available today for many applications because of its low scattering and absorption, large refractive index modulation capacity, moderate exposure, high resolution, and other unique capabilities such as reprocessibility. Holographic characteristics of dichromated gelatin are described on the basis of experimental data and several promising application areas of dichromated gelatin holograms are discussed.

Evaluation of the probability of error in detecting the state of a binary bit in a data page reconstruction is reported. Experiments are described that determine the effect on the probability of error of rotating the reading beam about the Braga angle. This effect is examined for volume Fourier transform holograms in iron-doped lithium niobate. Results are presented for three different diffraction efficiencies, for the incoherent superposition of hologram pairs with three different angular separations and for Fourier transform holograms with three different displacements of the electro-optic crystal from the transform plane.

The Thermoplastic-Photoconductor medium on a transparent polyester tape base is the subject of this paper. Specific areas of interest that will be covered are the use of the tape for holographic and bit serial optical data recording. The tape functions at speeds up to 30 ips and we report this performance with projections to 60 ips on both 4 mil and 1 mil substrates. High contrast reconstructions of 1-D holograms recorded on the moving tape (at a carrier frequency of 800 LP/mm) are shown as well as bit serial data bits (3 μm bits on 7 μm center) with a readout S/N of 10:1.

The main features of a non-volatile, optical storage scheme which has the potential of storing on the order of thousands of data bits in the frequency domain (f) are described. The frequency dimension can be used independently of the two or three spatial dimensions thus creating a three dimensional x,y,f- or a four dimensional holographic storage scheme. This frequency multiplexing method can, in principle, lead to an increase in storage densities of two or three orders of magnitude. The described scheme is based on a method of laser photochemistry which is known as photochemical hole burning (PHB); its limitations lie in the nature of the low temperature photochemistry as will be discussed in some detail.

This paper has five purposes. First, to focus upon the key relationships that bound the technology choices for large, archival, digital storage devices; second, to identify the motivations for selecting the optical technology for a petabit-exabit level storage system (1015-1018 bits); third, to present a generic example and a specific implementation of a terabit-level optical storage device; fourth, to characterize the global design space constraints that will allow one to build a tecnology-limited optical store; and fifth, to sketch the outline of the BYTERON concept, a wideband 1016-1017 bit optical store and contrast its performance to that of a current optical store.

This paper reviews the optical principles upon which ultra-high resolution optical readout is predicated. Several different recorded geometries and optical readout techniques are discussed. The discussion includes a description of focus and track sensing techniques and photodetector configurations. As an example of these optical principles the application of optical information retrieval to the quality control and monitoring of electromechanically mastered VideoDiscs is detailed and optical-readout performance assessed.

The Lawrence Livermore Laboratory has had a decade of experience with the use of an archival computer-storage facility. Along with other Department of Energy laboratories, we see a continuing need for such capabilities. This need arises from the increased speeds of large computers and from their proliferation as well. Surveys of potential sources suggest that the soonest and likeliest possibilities for developing an archival storage system are offered by optical technologies. The requirements for such a facility are briefly discussed.

This saper describes a microimage storage and retrieval system with application to fingerprint records. This system was developed to expedite the fingerprint identification process for the New York State Department of Criminal Justice based on their four million fingerprint card file. This unique system has successfully captured, recovered, processed, and displayed fingerprint images for comparison identification by skilled analysts.

An optical disc system is being developed to address wideband archival digital data storage applications. This system, which has the capability for "instant" playback (i.e., within a few bit periods), is configured about an RCA-proprietary "trilayer" optical disc structure. The optical disc has the potential for achieving high packing density (1011 bits/disc) and high data transfer rates. This paper discusses the RCA approach and experimental results of the implementation of an error detection and correction (EDAC) scheme and of a 50 Mb/s delay modulation encoding channel. The EDAC has been simulated using a computer program with raw data obtained from a readout of an optical disc at 40 Mb/s. The codes employed are described. The EDAC chosen gives about 100-to-1 improvement in error rates for 10 percent overhead. The 50 Mb/s delay modulation (DM) channel implementation will be described. Waveforms for the encoding and decoding signals are shown, along with a description of the logic circuitry used. Approaches for interfacing the encode/decode circuitry with external data equipment are also considered. The discussion of the DM channel includes a description of the synchronization and data identification signals and the manner in which they are serialized and inserted into the data stream.