It was just 20 years ago that A. B. VanderLugt, "Bud" to his friends and associates, had the flash of insight to realize that holography could be applied to the problem of recording the complex conjugate of a function, and thereby render possible the implementation of the correlation of two arbitrary functions optically. It is with his achievement in mind that we dedicate this special issue of Optical Engineering to the subject of Optical Pattern Recognition.

During a recent trip I waited for my suitcase after a heavily loaded flight from Atlanta to Los Angeles. As the various baggage items appeared on the conveyor, I reflected on the general problem of pattern recognition. The bags came in various shapes, colors, sizes, and orientations: my eyes casually scanned them as they appeared, and I found that I could quickly dismiss those that did not match my mental image of my own bag. I immediately recognized my bag when it appeared (it does have some distinctive scars after years of service) and resumed my trip to the hotel. While I was waiting, I also found time to observe the people gathered in the baggage area to see if there was anyone I recognized. Again, I was reminded of how subtle differences in facial expression, posture, and general body style allow us to instantly recognize people we know.

A polychromatic correlation detection technique by spectral-spatial matched filtering is described. This technique utilizes a diffraction grating and three collinear coherent sources. The technique for large capacity spatially multiplexing matched filter synthesis is also described. This technique offers true color signal detection, which is very suitable for color image recognition and identification. We also propose a real-time polychromatic image correlator for incoherent imaging. Several interesting experimental demonstrations of this color signal correlation detection scheme are provided. We note that this color signal detection technique will offer a wide range of applications for real-time color image recognition and identification, which is a step closer to the actual human perception system.

Computer simulations of the scaling and rotation sensitivity of a phase-only filter were performed, showing that it is much more sensitive to such input variations than is a classical matched filter. Values of the peak correlation spot power versus both rotation angle and scale factor are presented for both filter types. Several theorems are derived for calculating the optical efficiency of any filter in both input space and frequency space.

The use of a binary, magneto-optic spatial light modulator as an input device and as a spatial filter in a VanderLugt correlator is investigated. The statistics of the correlation that is obtained when the input image or the spatial filter is thresholded are estimated. Optical correlation using the magneto-optic device at the input and Fourier planes of a VanderLugt correlator is demonstrated experimentally.

Matched filters with signal-to-noise ratios that are space-invariant and rotation-invariant with respect to the target have been developed. Our approach has been to extract from the target one or more circular harmonic components and to use a filter matched to these components. The technique has been implemented both digitally and with an optical processor using com-puter-generated holograms. To discriminate between objects that are almost similar, a pseudo-object is used to generate the matched filter. The technique was introduced previously as an ad hoc method; a more systematic approach based on system constraints suggests the existence of other rotation-invariant linear filters.

When objects must be identified from distorted imagery, a choice must be made between feature sets that are invariant to the distortion and those that are not. Sets of invariants almost always contain less information, resulting in classification error rates that are higher under distortion-free conditions, but which are no larger when distortion is present. The choice can be evaluated by calculating error rates as a function of the eigenvalues of the correlation matrix, noise, number of classes, and a distortion parameter. An example of this evaluation is given by comparing identification of ships by using a subtraction correlator and moment features. The distortion parameter is scale, to which the correlator is sensitive and the moment comparison is invariant.

Synthetic discriminant functions (SDFs) allow distortion invariance to be achieved in optical correlators, thus making such systems more practical. The synthesis of four different types of SDFs is reviewed. Their performance in extensive projection tests on 144 images is presented, together with initial performance tests in the presence of noise. This distortion-invariant and shift-invariant pattern recognition algorithm can also be implemented digitally.

The optical implementation of geometrical shape and synthetic discriminant function matched filters is computer modeled. The filter implementation utilizes the Allebach-Keegan computer-generated hologram algorithm. Signal-to-noise and efficiency measurements were made on the resultant correlation planes.

The Hotelling trace criterion (HTC) is useful for feature extraction so that multiclasses of statistical images can be separated by maximizing the between-class differences while minimizing the within-class variations. Optical implementation of the HTC has been successful by utilizing computer-generated spatial filters and a coded-phase processor. A simplified method of calculating the HTC discriminant functions from large-dimensional images by a small computer is also described. This method is useful when the within-class variation can be approximated by a covariance matrix of low rank.

As the first step in investigating uses of the Wigner distribution function (WD) for pattern recognition applications, we consider its use in a simple binary detection problem. Using the output signal-to-noise ratio as a performance measure, we show that WD based methods provide similar performance to the conventional cross-correlator. We show that, by proper extension of the known reference signal, the computation involved in WD based methods can be reduced. Simulation results are provided to illustrate our theoretical conclusions.

The Wigner distribution function (WDF), a simultaneous coordinate and frequency representation of a signal, has properties useful in pattern recognition. Because the WDF is computationally demanding, its use is not usually appropriate in digital processing. Optical schemes have been developed to compute the WDF for one-dimensional (1 -D) signals, often using acousto-optic signal transducers. Some recent work has demonstrated the computation of two-dimensional (2-D) slices of the four-dimensional (4-D) WDF of a 2-D input transparency. In this latter case, the required 2-D Fourier transformation is performed by coherent optics. We demonstrate that computation of the WDF of real 2-D signals is susceptible to Radon transform solution. The 2-D operation is reduced to a series of 1 -D operations on the line-integral projections. The required projection data are produced optically, and the Fourier transformation is performed by efficient 1 -D processors (surface acoustic wave filters) by means of the chirp-transform algorithm. The resultant output gives 1 -D slices through the 4-D WDF nearly in real time, and the computation is not restricted to coherently illuminated transparencies. This approach may be useful in distinguishing patterns with known texture direction. The optical setup is easily modified to produce the cross-Wigner distribution function, a special case of the complex, or windowed, spectrogram.

A linear coherent optical technique for restoration of images altered by a multiplicative periodic degradation is considered. By proper choice of a multiplicative grating, restoration can be performed using conventional linear coherent optics techniques. Specific attention is given to the case of continuous sampling in which the image is periodically set to zero in strips.

That the laser has had an enormous impact on modern spectroscopy is history. However, this impact was long sequestered in academic pursuits, where a myriad of "new" light scattering phenomena were observed. Most of these were previously unobserved light scattering effects due to second- and third-order terms in molecular polarizability. Toward the end of the last decade, the field of laser spectroscopy reached a maturity, and numerous laser spectroscopic techniques began finding practical application.

Twelve laser-excited atomic fluorescence methods suitable for absolute temperature measurements in flames and other atomic and/or ionic reservoirs are reviewed and summarized. The different characteristics of the techniques are discussed. Several important parameters that need to be evaluated experimentally and the assumptions that need to be made in order to obtain meaningful temperature data from the selected method are emphasized.

Photothermal refraction is a novel thermo-optical technique that measures absorbance from localized regions within larger samples. In this technique, a modulated pump laser beam and a coplanar continuous-wave probe laser beam cross at right angles within the sample. Absorbance of the pump laser beam generates a time-varying localized temperature rise within the sample. This heated region acts as a thermal cylindrical lens that periodically defocuses the probe beam and results in a change in the probe beam center intensity. Since the pump and probe beams interact only in their intersection volume, a localized measure of sample absorbance is produced. A high spatial resolution image of sample absorbance may be made by recording the photothermal refraction signal as the sample is scanned through the intersection of the laser beams.

Large mirrors appear to be limited in resolution by their medium scale defects (mirror surface "ripples" of small amplitude and of 4 to 12 cycles per diameter). The ultraviolet resolutions of the Pic du Midi (PDM) 2 m mirror, the Canada-France-Hawaii (CFH) 3.6 m mirror, and the Space Telescope (ST) 2.4 m mirror have been studied. Hartmann screen tests were used to determine the wavefront error of the PDM and CFH mirrors, while only semiquantitative reports were used in the case of the ST mirror. Point spread function and modulation transfer function were determined by two-dimensional fast Fourier transform analysis, a necessary technique in the ultraviolet, where the small phase defects of mirrors are no longer negligible (and also because the mirrors' irregularities do not possess a particular arrangement or a given symmetry). If the three mirrors appear to be nearly limited by diffraction in the visible (except the CFH), in the ultraviolet at Lyman a 121.6 nm the resolution is inferior by more than a factor of 10 to the diffraction-limited resolution. While the ultra-violet wavelength range and very high resolution (0.014 arcsec with a diffrac-tion-limited Space Telescope at Lyman a) are highly desirable, such a limitation imposed by mirror surface quality has to be mentioned. The effects of the correlation length, amplitude of defects, and Hartmann screen sampling are presented, as well as some comments on the inadequacy in the ultraviolet of analytical methods compared to two-dimensional numerical simulations.

By using three noncoplanar coherent laser beams intersecting one another at angles larger than 30°, three sets of interferometric gratings are generated. By placing a specimen coated with a photosensitive material inside this grating field and exposing it before and after load, one can obtain three distinctive moire fringe patterns, from which principal surface strains can be calculated by using the strain rosette formula.

Analyses are made of technological progress, using four technical areas as case studies. The progress in an area is measured by the cumulative number of "valuable events" in that area, where an event is defined as a new idea, a new structure, a new material, or anything new that is reported, developed, or invented. The four areas studied are (1 ) transverse mode stabilization structures of the semiconductor laser, (2) excimer laser gas materials, (3) solid-state laser materials, and (4) optical storage media. It is found that the rates of change in the cumulative event numbers are well described by the Verhulst equation and that progress time histories are well represented by sequences of logistic curves. It is found that most technological progress is characterized by a fast growth period of about 10 years with a time constant of 1 to 3 years. This first growth phase is followed either by saturation or by a smaller secondary growth phase.

Current work with the cavity phase shift (CAPS) method concentrates on hydrogen fluoride laser wavelengths between 2.6 and 3.0 um. This method and continued advancements in experimental techniques are described. With use of a cavity dither the accuracy of reflectance measurements has been improved over previous work to ±0.0004 for reflectances around 0.995. Results of measurements on dielectric-coated elements are reported, as are studies on metallic monolayers placed on nontransmitting substrates. The required CAPS configuration in these cases is a three-mirror arrangement permitting angle-of-incidence variation. In this mode the CAPS method becomes a verification technique for optical coatings on particular resonator designs.

Measurement of deformation or shape differences of two objects is impossible in usual holographic interferometry. Difference holographic interferometry (DHI) overcomes the difficulties by producing a comparing technique for measurement of displacement fields and shapes of different objects and by being powerful in the case of large deformations. The main feature of DHI is that holographically recorded waves are used for illumination by wavefront reversal. It works without the need to consider different microstructures of object surfaces. A few realizations of the technique are presented.

A multidirectional laser beam scanning method using holographic zone plates is described. To increase the number of scanning directions, not only is a hologram disk rotated at fixed speed (col ), but a laser beam incident on the disk is reversely rotated at another speed (w2). The focus of the laser beam diffracted by the hologram disk draws multidirectional scanning lines. The number of scanning directions (n) is equal to (w1 /c02) + 1. Experimentally, the effectiveness of this method is demonstrated for nineteen scanning directions. This scan method is applied to a point-of-sale (POS) scanner to read a non-moving bar-code symbol.

Shading correction of image sensors is a very important operation in visual inspection applications. The generation and application of shading cor-rection with and without imaging system response linearization is discussed and demonstrated. The linearization is achieved by subtracting the dark current image and by applying a look-up table operation. The content of the look-up table is obtained by fitting an analytical function to the measured system response. The linearization enables the shading correction to be a linear operation. The quality of the linearization and the shading correction is evaluated using statistical parameters of the processed images. The results show a substantial decrease in the standard deviation of the gray level distribution of uniform reference images.

Image restoration filters are known to improve the subjective quality of a picture. However, it is not evident that the filters would facilitate further machine processing such as machine classification. In this paper, several image restoration filters are evaluated and compared based on their perfor-mances in machine classification under various blur and noise conditions to assess their usefulness for automatic classification. An experiment has been specifically designed and conducted for this purpose. Appropriate performance measures are derived. The results show that, under certain combinations of blur and noise conditions, the filters do improve machine classification.

The tolerance on decentering and tilt of the secondary mirror of a Cassegrain/Nasmyth telescope is considerably broadened when the zero-coma condition is applied, the limit being set by the allowable residual coma and tilt of the focal surface. The dependency of field tilt on decollimation is defined. A simple set of steps for effective collimation of a telescope is outlined, along with practical engineering recommendations concerning the Ritchey-Chretien (R-C) astigmatism corrector and focal surface subsystem. It is shown that when the proper engineering degrees of freedom are provided, essentially ideal R-C imagery can be obtained over the original optical design field, even in the presence of relatively large angular errors in the optical axes of the primary and secondary mirrors. The specific example given is the case of the proposed University of Texas 7.6 m telescope.

Matters relevant to preparing an optical flatness standard are discussed, such as selection of the material, surface preparation, bulk viscoelastic deformation, thermoelastic deformation, refractive errors, reduction of interferograms of plates examined in pairs, absolute calibration by intercomparison of two flats examined in two orientations, and the combination of several contours to obtain the figure of the plate. The contours are given as Fourier sine series expansions, and a technique for obtaining the coefficients is described.

A design approach for a camera to be used with the Space Telescope is given. Camera optics relay the system pupil onto an annular Gaussian ring apodizing mask to control diffracted light. Image plane intensity distributions for both one- and two-dimensional models of ripple on the primary mirror were calculated. Calculations using ripple amplitudes between X/20 and X/200 with spatial correlations of the ripple across the primary mirror between 0.2 and 2.0 cm indicate that the detection of an object 109 times fainter than a bright source in the field is possible. We thus conclude that detection of a Jovian-type planet in orbit about a Centauri with a camera on the Space Telescope may be possible.

The purpose of this letter is to point out a disagreement with a recent paper 1 on the above subject. For an opaque rectangular object having a width 2a and a height 2b with its sides located parallel to the axes (Fig. 1), the intensity distribution in the (x , y) plane is given by the following expression:

Every technical paper satisfies one or more purposes, including - information dissemination - claim staking - self-promotion - marketing - conference presentations - competition with one's colleagues and rivals - pressure to "publish or perish."