The Aerospace model of the infrared radiation of the zodiacal light has been generalized to include the effects of scattered sunlight. The model has been compared with existing data on scattered visible light in order to test whether or not the infrared data are compatible with the visible data, i.e., can both the infrared and visible data be explained simultaneously by a single self-consistent model of the interplanetary dust? Both the IR emission and the visible scattering calculations are based on a Mie calculation, with the optical constants of the interplanetary dust being input parameters. The size and space distribution of the dust particles are explicitly parameterized; the model numerically integrates the results of the Mie calculations over the size distribution and along any chosen line of sight. The results show that the IR and visible data are in fact compatible. Both sets of data can be matched by the model by adjusting the continuous opacity within reasonable bounds. This is equivalent to adjusting the albedo of the particles. Furthermore, the size distribution function is very tightly constrained by comparing the predicted visible polarization of the model with existing data. This constraint leads to the conclusion that the IR emission cannot exhibit emission features.

Background suppression techniques and algorithms for on-board processing applications to earth-looking and deep space IR detection and tracking systems are presented. Modeling considerations of the background and sensor system noise sources are addressed. Some current research and development efforts in signal extraction and background suppression techniques, and computational aspects are briefly discussed.

Scaling laws are derived for the calculation of clutter leakage in sensors which display mixed jitter and drift relative to a background characterized by a two-dimensional power spectral density which can be approximated by an inverse power law. These scaling laws are applied in the construction of a series of nomographs permitting rapid estimation of clutter leakage for arbitrary detector footprints, drift rates, jitter statistics and integration times, for nth order frame differencing with various PSD power laws. Application of the nomographs to staring sensors with temporal n-th order difference filters is explained, and limited application to spatially-filtered staring sensors and scanning sensors is discussed. The extension of the nomograph set to cover other PSD slopes and viewing angles is discussed. The paper then concludes with a general discussion of relative accuracies and utilities of PSD-based clutter leakage estimation versus simulation methods.

A new generation of solar arrays is in production utilizing technology capable of satisfying the large (possibly flexible) solar array requirements of developing space programs, such as Milstar and Space Station. These programs have turned to infrared technology to satisfy five of their production needs. First, in the welding of solar cells to their interconnect circuitry (module production) a high speed, infrared thermal controller is utilized to stop the welding process when a preset temperature at the weld area is attained. Second, after an actual weld is made, it is inspected with a laser-based, infrared stress sensor to determine its presence and, potentially, its quality (strength). Third, these welded solar cell modules are inspected for cracks and abnormal stresses with an infrared microscope before their integration into panels (arrays). Fourth, the bond quality of solar cell panel substrates (skin plies and skin to inner honeycomb) and of the individual solar cells to the substrate are reliably determined by infrared thermographic techniques. Last, the final arrays and panels are inspected for cracks, and weld joint presence and quality utilizing infrared sensors with stress visualizing and quantizing capabilities. In order to efficiently perform these tasks, large automated (computer-based) systems position the sensors or the workpiece, perform measurements and calculations, and evaluate the results for process control.

Ten years ago infrared imaging systems available on the commercial market had reached a point in their development where accuracy, speed, thermal sensitivity and spatial resolution were sufficient to meet the vast majority of measurement requirements. They were severely limited in application potential, however, because the images produced by even the highest performing systems appeared on oscilloscope displays or Polaroid prints with no further image or data analysis offered. The development of the personal desk-top computer and its marriage to the commercial infrared imager was the key to an applications explosion for these systems. The addition of compatible videocassette recorders added even more to their versatility. This paper will trace the development of commercial infrared thermal imaging systems since the advent of the personal computer, provide an overview of some of the more outstanding features available today and make some projections into future capabilities.

In this paper the design and development of a small-aperture imaging system used for surveillance of vehicular traffic at night is discussed. Such a sensor may have applicability to the safeguarding of a secure facility, where it is important to monitor vehicle activity near the facility perimeter. The sensing element is a linear pyroelectric focal plane array. Motion of a target vehicle provides for scanning in the horizontal direction, and negates the need for a chopper to produce a differential IR signal. Also presented are various digital image processing procedures that are employed to enhance the collected image data, since the images produced by such an array are qualitatively quite different than those obtained in the visible portion of the spectrum. The array consists of 32 lithium tantalate (LiTaO3) elements spaced on 15 mil centers on a single crystal, with element dimensions of 13 by 32 mils. Thirty-two individual on-chip hybrid detector pre-amplifiers reduce the large switching noise that would accompany post-multiplexer amplification. Processed images collected with this prototype system are shown.

A radiometric model has been developed for modeling the effects of self emission, reflection of downwelled radiance, atmospheric transmission and path radiance on the radiance reaching an airborne or satellite sensing system. The model, which incorporates LOWTRAN 6, includes angular emissivity effects and will account for either specular or diffuse reflectance from targets and backgrounds. The model has been refined to facilitate its use in generating primitive synthetic images to study LWIR phenomena. In particular, various types of sensor related noise have been included in the imaging process. Images are presented simulating the effects of target-to-background contrast as a function of view angle, range, emissivity variation, and atmospheric effects. The potential for use of LWIR models in sensor and algorithm development is discussed.

Image tracking algorithms, particularly, pixel level based frame to frame correlation techniques need realistic dynamic scene sequences in order to evaluate their performance and to know whether the algorithms are converging or not. The cost effective way of obtaining these images is only by means of understanding the various constituents of the system, their relationship through a mathematical model and then simulating the required scenario. This paper describes such a scheme - IR Image Simulation (IRIS) when both the target and the sensor are in motion. Details of mathematical model of a tank like target are mentioned in this paper. Development of software packages for simulating background, receiver optics, atmosphere etc., are realized based on the mathematical models which are taken from the open literature. A methodology for real time usage of this system is also briefly mentioned in this paper. Examples of computer simulated images are also presented.

The spatial and statistical properties of the radiance over a scene of natural terrain is of importance while a target detection algorithm is developed [1]. In geological studies the use of I.R. imaging is in-creasing and the understanding of the image is of significance L2]. One dimensional measurements and anal-ysis of the thermal image of natural terrain was performed by Itakura and his coworkers [3]. In this study a model was proposed, describing the statistics of the radiance over a given scene and the one dimensional power spectra of it. The main claims were generally validated by later work of the ERIM group [4,5].

The equipment and data analysis used by a field laboratory system to measure atmospheric transmittance spectra are described. Measurements have been made at distances up to 44 Km in the 3-5μm region and up to 24 Km in the 8-12 μm region. The results agree well with the latest version of the LOWTRAN computer code.

The Minimum Resolvable Temperature Difference (MRTD) is a classic performance parameter, used by manufacturers of thermal imaging systems (FLIR's) in the characterization and final testing of such devices. The Johnson criterion [1] relates the MRTD to practical characteristics, such as the ability of the FLIR to resolve objects, as function of distance. However, the MRTD measurement suffers from several drawbacks,e.g. i) it is subjective, ii) it fails to account for complex real life situations, such as background and foreground clutter, object shape and non-uniformities of temperature and emissivity. This paper describes a new type of transparency which has been developed to simulate real infrared scenes. The transparency consists of a half-tone mesh of variable size holes which, when placed in front of a blackbody, simulates a two dimensional radiant emittance pattern. The pattern itself can be generated from a thermal image of any real object as seen by a high performance FLIR, from an artificially computer generated image, or from a combination of the two. Examples of thermoscenes and measurements of radiation distribution over a simple pattern are shown. Potential applications of the thermoscenes are also discussed.

The models currently used for line-of-sight (LOS) MTFs associated with FLIR platform instabilities do not account well for the frequency content of the vibrations, the ability of the brain to "track" certain motions, or for the frame-to-frame integration. This paper presents some first results from LOS MTF measurements carried out with fast-scanning FLIRs subjected to sinusoidal and random vibrations. Only the direction of continuous scan has been considered at this stage. Our results show that the current models predict LOS MTFs which are too pessimistic. This is particularly apparent in the case of an observer-operated FLIR for the important range-determining middle spatial frequencies. The standard theory has been extended to account for frame-to-frame integration in systems which do not include an observer. This theory is used as a basis for discussion of all measured results, including those involving a human observer.

Large single crystals, up to 8" diameter, of high quality Optical Germanium, have been grown by the Czochralski technique. Post-growth thermal treatment improves the optical homogeneity and reduces optical losses, as shown by refractive index gradients and MTF measurements. A new approach for the piecewise combination of interferograms as well as a polychromatic treatment of MTF, are presented.

For the purpose of current leakage analysis in p/n junction narrow gap semiconductor diodes, the effect of hydrogen ion implantation and hydrogen plasma on InSb photovoltaic I-V diode characteristics was studied. The use of hydrogen discharge technique for semiconductor surface passivation just before or during the plasma CVD deposition of Si02 indicates that it is possible to obtain control of the flat band voltage parameter. Nevertheless, measurements of C-V hysteresis loops of MIS devices and I-V leakage measurements on diodes that had the same hydrogen passivation treatment as the MIS devices, do not show control of these parameters. For current leakage location identification we have carried out experiments with gate-controlled diodes (G.C.D) whose gate is located both on the mesa slope and at various distances from the slope, on both sides of the junction. In the electrical experiments with G.C.D, the device was operated either under external stress (compressive or tensile) or under stress-free conditions. A quantitative model was developed for calculation of the current leakage, including the I-V gate- controlled diode characteristics. The model was based on an algebraic relationship between the gate voltage and the charge density near the oxide and the p/n junction and it has the potential to include different current leakage mechanisms. The model was confirmed experimentally, and an equivalent defect charge density in the layer between the oxide and the semiconductor, near the p/n junction, was obtained by analyzing either gate voltage or external stress. Using the parameters of gate voltage and external stress, we can induce various kinds of I-V leakage characteristics in a leak-free InSb diode, including negative resistance to forward bias. However, similar methods were not successful in inducing negative resistance characteristics in Hgl_x Cdx Te (MCT) photovoltaic p/n+ diodes. Deep energy-state density in the energy gap due to material defects or process-induced defects, may be responsible for the high excess currents observed. A discussion of the G.C.D.-induced current leakage characteristics as indicators for high quality substrate material in p/n diode fabrication is given.

This report covers a life demonstration test which was carried out during July 86 - March 87 as part of Ricor Ltd. product improvement program. 8 Units STRIX 1/4 W Split Stirling Cryocoolers were subjected to test profiles according to NVEOL Specs. 132-28A050127 A - 4 units, and IDF specs A001/86 - 4 units. Ambient temperature span in both profiles was -32 .. +52°C. The IDF specs. include random vibrations for 10 hours every 250 hours. The test was witnessed by and reported to the IDF test and evaluation unit and El-Op Ltd. QA department. Meantime to failure (MTTF) observed in test was 2020 hours, standard deviation of 563 hours. About half of the failures were assumed to be the result of gaseous contamination. Compressor and motor MTTF were well above 2000 hours. Operation in a closed loop temperature control mode by power conditioner/temperature controller module EM4138 can extend cooler reliability by about 30 %.

Infrared sensors can detect objects by means of their thermal radiation contrast with the background. Systems to warn of the approach of threats, such as personnel, missiles etc., can be built using these sensors. The requirements and design choices of warning systems differ in many respects from those of thermal imaging systems. This paper describes two IR systems which were developed for warning purposes - the PWS, designed to detect intruding personnel and SPIRTAS, a system to protect ships by warning of the presence or approach of surface craft, airplanes and missiles.

The output signals from FL1R sensors suffer from many degradation mechanisms including high clutter noise, detection noise and strong dependence on weather and time of aquisition. The degradation mechanisms are more pronounced for ground terrain scenes, which are the cases we are considering. An automatic target detection system, must overcome these degradations. Degradations due to the target's own thermal signature distortions (scale, rotations, aspect angles, partially hidden parts and screening by smoke or dust) have to be tackled too. Thus, "automatic FLIR target detection" involves a massive computational burden. Here we introduce a system based on a highly parallel compact hybrid thermal imagery signal processor in which the main computational route is executed in an analog optical processor module. The input video signal, after electronical on-line preprocessing is introduced via a spatial light modulator to a coherent optical processor. The image is optically prefiltered using an appropriate synthetic filter, and then Fourier transformed- using a lens, in a 4f configuration. A transformation replicator is utilized in order to obtain several spatially separated replicas of the transform. Each one of the Fourier transforms is spatially multiplied by a different optically implemented algorithm and then optically inverse transformed. The output signal is collected by an array detector and is transfered to a digital post processor. Some of the optically implemented algorithms are used for background rejection. The other algorithms are used for the detection of regions of activity and targets by their thermal signatures. The detection algorithms were designed in order to achieve high detection rates with target scale invariance. Preliminary examinations of the overall processor have yielded high positive detection rates, and medium false alarm rates.

The experimental activities of the IR group at Israel Aircraft Industries (IAI) include the performance of IR signature measurements of aircraft in static ground as well as dynamic flight tests. Measuring capability exists in spectrometric and band modes. In support of measurements, signature modelling is performed. These include emissions from gaseous and solid-body sources, in the near, middle and far IR spectral wavebands. In support of related projects a test range has been developed which includes a jet engine simulator. This test facility has the capability to produce the IR radiation emitted by the hot metal surface and gaseous plume of a typical aircraft engine, at different power-level settings. These simulations are for low altitude aircraft, with flight speeds of up to 1 Mach. The IR radiation produced by the engine simulator is measured by a radiometric system, designed and constructed at IAI. The system includes five radiometers, operating simultaneously at 5 out of 13 different angles. A comprehensive base of computer codes at IAI is available and extends the understanding of the processes involved. These include the possibility of performing heat transfer calculations, jet engine wake temperature and chemical concentration distributions and atmospheric transmission considerations. In this presentation some of the results of radiometric measurements of the engine simulator will be compared to theoretical estimations of its IR signature.

Graded Index material (GRIN) is presently being used as a flexible means of problem solving in special visual optics applications. For the optical designer GRIN serves as a novel additional design parameter whose actual employment is limited only by the ability to produce the required index profile. The most widely used commercially available GRIN material for the visible region is known under the name of selfoc, a proprietary process of a Japanese firm. The high cost of IR transmitting materials and the complex process of aspheric element fabrication suggest that IR GRIN be used as a flexible and cheaper implementation alternative. So far, nothing but theoretical work has been carried out concerning the use of GRIN elements in the IR region. To facilitate GRIN elements, three basic capabilities are required: GRIN fabrication Index profile measurement Utilization of GRIN in the optical design An original method for the fabrication of graded index elements from IR transmitting chalcogenide glasses of different compositions has been developed. The preparation method and some application of IR GRIN glasses are presented. A sample IR system employing GRIN element will be presented as well. There should be few words on the glass employed. The compositions and profiles were measured by the EDAX technique, radioactive tracing and optical methods. The diffusion parameters were calculated for Se in As2Se3 glass and are the following: Diffusion coefficient Do = 3.65 10-6cm2/sec. Activation energy Q = 3.95 k.cal/mol degree. The results will be discussed.

Large, high quality, single crystals of sapphire and calcium fluoride are routinely grown at the NRCN laboratories via the Gradient Solidification Method (GSM). This method was initially developed for the growth of dome-shaped sapphire crystals and then also applied to the production of bulk cylindrical crystalline boules. The diameters of the currently grown crystals are 150 mm and 85 mm for the sapphire and calcium fluoride respectively. The development of growth procedures for larger diameter sapphire crystals (up to 12") is underway. The growth processes were investigated and crystal imperfections (i.e. grain boundaries, twins, scattering centers, etc.) were correlated with various growth conditions. The optical and mechanical properties of the grown crystals were measured on fabricated components.

A laboratory simulator for dynamic infrared scenes was built and used for evaluation of a FLIR tracker. Background and targets were simulated by separate fine mesh patterns. These are made by a special process enabling one to simulate real or synthetic (computer generated) infrared scenes. The set-up can be controlled to simulate target movements against a background in two degrees of freedom. The set-up also simulates independent tracker movements. The problems of superpositioning the target on the background in such a way that the target will not appear semitransparent, was achieved by meticulous set-up design. The simulator was used to measure the dependence of the performance of a FLIP tracker on: clutter, background temperature, target contrast, target movements, target range, and line of sight movement and vibrations. The paper describes set-up design considerations, target and background model fabrication methods and experimental test data.

0.1eV HgCdTe photoconductive detectors with optical masks have been developed to improve spatial resolution and sensitivity. Measurements on carrier lifetime, responsivity and noise were made for the newly developed detectors as well as ordinary ones. The following results were obtained. (1) Measured responsivity for the new detectors was 2-3 times as high as that for the ordinary ones. (2) NEP for the new detectors showed better performance in wide bias current range. Finally, calculations on carrier lifetime, responsivity and noise were carried out by solving a 3-dim. transport equation. The calculated values agreed well with the measurements.

The attenuation of IR radiation by the raindrop particles can be calculated if the rain drop size distribution is known. Number of models exists to predict raindrop size distribution, among them Marshall-Palmer and Joss-Waldrogel models. Recently, 2 independent field measurements during rain were performed in U.S., that showed the existence of the particles in the 10pm size range, which are not predicted by the existing models. First field measurement program involved direct measurements of the size distribution using Knollenberg PMS system and simultaneous measurements of IR and visible attenuation. Second field measurement program involved the measurements of the forward scattering of UV radiation by the particles using UV Forward Scatterometer. The results of 2 independent programs are presented and discussed. They indicate that the existing rain models will not predict correctly the attenuation in the thermal IR region.

The spectral background radiance and the spectral transmittance are experimentally investi-gated in 1.5 - 14.5 micron band and compared to LOWTRAN V model calculations. We use a BARNES spectroradiometer with CVF covering continuously 1.5 - 14.5 micron and two AGA Thermovision 782 IR scanners with three discrete filters in 3 - 5 micron band and 8 -12 micron band. Spectral and spatial IR data are available on-line by means of two personal computers: an HP9836 that drives the spectroradiometer and elaborates the data as a function of wave-length and an IBM PC-AT that drives the scanners and elaborates the spatial data in order to obtain pictures (512x512 pixels)displayed on a H.R. monitor. Our final goal is to identify the most important parameters affecting the discrimination of targets from background (spectral band choice, temperature, emissivity, and so on). To this end we measure also the IR energy emitted by a reference plate at different values of temperature and emissivity vs. various background previously investigated. In particular targets that are close to ambient temperature are considered. Emissivity gets estimated in our lab by means of a dedicated experimental set-up. This study is connected with a Piaggio R&D program in the field of IR coating and chemical reduction of IR signature for aircraft application.

The boresight alignment testing for the Sun sensor assembly on the Halogen Occultation Experiment (HALOE) is described. The Sun sensor assembly consists of three sensors that provide feedback signals for controlling dual axes gimbals. Two energy balancing silicon detectors are operated as wide-band sensors in the azimuth and elevation axes. The third sensor is a silicon photodiode array operated as a narrow-band sensor in the elevation axis. These sensors are mounted on a common Invar structure which is mounted to the HALOE telescope. A blackbody was used as the stimulating source to perform the initial boresight alignment and this was checked with a heliostat solar look and a direct solar look. These tests are explained with a comparison between each source used.

In recent years ion implantation has been used to develop MCT and other infrared photovoltaic detectors, However, the implanted ions usually have a nonuniform depth profile in the implanted layer, with the typical depth less than lμm, within which the concentration of the implanted species may change over 3 to 4 orders. Thus within the layer there is an electric field outside the space-charge region. The electric field can be estimated with its intensity as high as 1x102-1x103v/am at 77k. In the present work the effect of the electric field is studied to show its way to effect the performance of the infrared photovoltaic detectors. Using the formulae obtained in this work, the RoA, product, responsibility, noise and detectivity of n-on-p Hgo.8Cd0.2Te photovoltaic detector operating at 77k are calculated for 300k backgrond and 27πfield of view. The results show that they are all related to the electric field in the implanted layer. It is interesting that when the electric field direct to the surface of n-region, namely negative field, the RoAlv=0 product contributed by n-region can not be neglected and the responsibility R,. and detectivity Dℓ, decrease. In the case of positive electric field, responsibility RA and the noise Vn both increase while the Dℓ increases slightly. Our calculation suggests that the proper technology process should be taken in manufacturing the MCT n-on-p infrared photovoltaic detector by use of ion implantation to eliminate the negative field in the n-region in order to avoid the harmful effect. The calculation is performed to show how the electric field in the ion implanted layer on Hg0.8Cd0.2Te to effect the performance of Hg0.8Cd0.2,Te photovoltaic detectors. The result shows that a reverse electric field is unfavorable for the performance of the photovoltaic detectors.

Data reduction techniques have been implemented recently to obtain infrared spectra of astronomical objects from raw data obtained with the mosaic detector array and prism slit spectrometer used for groundbased telescope astronomical observations. During the course of the analysis, several instrumental peculiarities presented significant complications that were ultimately addressed. The techniques for flat fielding a slit spectrometer cannot follow exactly those used for an imaging camera, but a limited flat fielding approach can be used in conjunction with measurements of known calibration sources to calibrate the entire array. The end product described here allows a pipeline application of computer software to raw data frames that results in astronomical spectra independent of atmospheric and instrumental attenuation.

Even under the worst conditions, a surprisingly large fraction of practical ICBM trajectories from launch sites in the Soviet Union to targets in the U.S. are sunlit for a signifi-cant portion of the trajectories' midcourse. Many trajectories are never wholly within the night shadow of the earth. Graphs and tables are presented to illustrate this. The impli-cation is that SDI space sensor systems can advantageously use visible light cameras. Since visible light cameras are much simpler and less costly than infrared cameras, this can both simplify some systems, reduce their weight, and reduce their cost. This should be of importance especially for the earliest deployed SDI systems.

An IR transparent, but microwave opaque, conductive mesh structure is developed to shield an infrared (IR) system from intense electromagnetic (EM) interference in the microwave range . To prevent potentially damaging EM radiation from penetrating or coupling into the IR system, the IR scanner is enclosed in a Faraday cage. One implication of this IR camera shielding technique is an ability to acquire accurate IR data in any noisy EM environment. A second application of the conductive mesh involves the investigation of EM fields in a copper cylinder, closed at both ends, constructed with a copper mesh section incorporated at one end of the cylinder. Using several positions for the detection material, three-dimensional field profiles are obtained and the results are presented.

A remote temperature sensor has been developed to measure the temperature of a metal alloy in microgravity conditions, on board an automatic and sealed experimental station in the Space Shuttle. The device is composed of an optical system, that allows thermal radiation from the hot body to pass through an orifice in the oven, and focus onto an infrared detector whose output photocurrent is a function of temperature. The experiment is fully microcomputer controlled, and temperature data is both, utilized to control the heating path, and stored in the memory for postflight analysis. In this paper, design, construction, calibration, testing and performance of the system are described.

This paper examines the potential benefits of using focal plane arrays in I-R detection systems. Two basic modes of operation are described together with their applicability to both the medium and long wavelength bands. We contrast the advantages and problems of two different readout schemes and consider their interaction with off focal plane electronics. We review theoretical ultimate limitations to performance and present more realistic theory predicting practically achievable results. Important array parameters such as uniformity, stability and dynamic range are considered together with constraints imposed by drive electronics, optics, cryogenic systems etc. Based on these considerations, our approach has been to confine the focal plane electronics to the functions of integration and mutliplexing, performing more power consuming tasks off the focal plane.

High performance focal plane arrays need associated electronic drive and signal conditioning circuits. 1Ne review the requirements of different array types and discuss a flexible, modular system which may be used with all the arrays described in the preceeding paper. This system can be used as a research tool to investigate the operation and performance of arrays. It can be used in its present form in some applications but has also been designed to be miniaturised where limitations of space or mass dictate this. The system has been designed to allow maximum flexibility of the operating conditions, consistent with a realistic implementation. To achieve this, most of the parameters can be programmed from a desk-top computer, which also facilitates storage and retrieval of information. However, in real applications the operating parameters may be easily stored in firmware. In addition to the basic drive and signal conditioning functions, the electronics package can also control additional functions such as micro-scanning, "snap shot" mode of operation and conventional mirror scanners for either scanned linear or block scanning.

This paper describes the circuit techniques used to implement a flexible set of correction, control and drive modules for use with a range of photovoltaic, Infrared arrays. The system consists of a set of modules inter-connected via both data and parameter busses. This structure allows selection of the appropriate set of modules for any desired IR array, whilst maintaining the same data and control formats. This approach allows several units to be connected in parallel to process many pixels, either from many conventional (-104 detectors) arrays or from a single larger array. We examine the problems of interfacing both high speed clock waveforms and low level signals to the dewar. The correction processes are described together with the importance of a common output format suitable for a wide range of users. The format has been selected to allow the future use of new generations of monolithic digital signal processors. The flexibility of this electronics allows it to be extended to form a programmable assessment system capable of measuring Infrared arrays under realistic operating conditions. Finally we present results obtained from this system in conjunction with a variety of arrays.

This paper presents a computer program for simulation of an infrared focal plane array. Standard equations are used to support a menu driven program developed for an IBM personal computer. The terms and equations for each section are presented and samples of actual screen displays of a currently available device are also included. The program is intended to provide the user with a better capability to understand and to study the tradeoffs of fabrication parameters versus the focal plane array performance (i.e. CTE, both spatial and temporal dynamic range, MTF, and noise) used for an optical sensor system analysis. Only surface channel devices are considered in the simulation.

We describe the operation, construction, and performance of an infrared focal plane array that utilizes platium silicide Schottky-barrier diodes on p-type silicon as the infrared sensor and has a MOS/CTD readout structure. The i28 X 128 element array is fabricated using standard integrated circuit grade silicon and NMOS processing. Associated with each photoelement there is a MOS multiplexer switch. Parallel connection of every multiplexer switch in a row allows a digital shift register to address one row of diodes at a time. The video signals are parallel transferred through common column video lines into two bucket-brigade device (BBD) charge transfer registers for signal readout. This readout structure has the advantages of high fill factor, low noise, and large charge handling capacity. In addition, the digital scan register and the BBD registers are less susceptible to freeze-out at temperatures below 40 kelvin than a buried-channel CCD.

This paper discusses the Navy program for infrared focal plane array (IR FPA) development and some problems involved in the development of high performance IR FPAs for several advanced infrared sensors. Emphasis is placed on specific IR FPA technologies currently being developed as part of the Navy Program. Issues involved in the determination of array specifications are demonstrated using a parametric analysis of performance predictions for an advanced staring sensor. The analysis includes consideration of system design parameters such as IFOV, FPA noise and uniformity, clutter rejection and signal filtering.

A 4-band X 4096-element IR-CCD has been developed with Schottky-barrier detectors operating at 80 K. The device is a monolithic quadruple-band image sensor in the short wavelength infrared (1.6 pm - 2.35 }gym) band. This imager has a stagger detector layout. The detector size is 10 μm X 10 μm and the effective fill factor is 100 %. A package for the SW IR-CCD has been also developed. The package is a SiC-Al203-SiC structure. The device is assembled with epoxy. The epoxy mounting does not degrade after 9000 times heat shock experiments (77 K - 333 K). The measured performance of the device includes the transfer inefficiency of the CCD, signal to noise ratio, dynamic range, uniformity and power dissipation of the device.

A catadioptric Cassegrain optical system is described that permits simultaneous operation as an infrared (IR) imager and a millimeter wave (MMW) radar through a common aperture. Separation of IR and MMW focal points is achieved by use of suitable dichroic coatings and substrate materials. We have evaluated our design by building a prototype and measuring its performance in the laboratory with particular attention given to isolate the degradation suffered by each mode due to the presence of the other. We present our test results and conclude from an analysis of the data that although our design is feasible, improvements are needed both in the design of the secondary element and in the technology for conductive dichroic coatings. We suggest a straightforward solution to the problem with the secondary element and present evidence to show that requirements for conductive coatings will soon be achieved.

Rapidly changing infrared scenes can be generated as a function of position using thermally produced irradiance. This can be done most conveniently by using a variable dwell-time, variable diameter, and scanning laser beam to deposit a predetermined amount of laser energy, as a function of position, on the target surface. An infrared target simulator has been designed and tested to perform within stringent temperature simulation requirements: heating rate of over 100 K/sec is achieved, and temperature gradients of 200 K/mm can be maintained. Problems with heat transfer between adjacent elements at different temperatures are solved by modeling the thermal layer on the target surface as a low-pass filter, which cuts off all the frequencies above the thermal cutoff frequency. When the system approach is taken to the infrared simulation design, the optical system limited performance results in the best realizable performance. In this case, the thermal layer on the target surface is designed so that the thermal cutoff frequency is above the optical cutoff frequency.

We have developed a highly sensitive, compact SO2 gas sensor that uses a PbSnTe laser. To increase its sensitivity, we used wedge-shaped windows that decrease etalon fringes and a SO2 absorption line that is little influenced by water vapor and that has a strong coefficient. The etalon-fringe intensity of the optical system was reduced to 1 x 10-4 and the SO2 absorption line selected is 1380.9 cm-1. To make the sensor compact, we developed single-mode lasers to eliminate the need for a monochromator, used a 30-cm multipath cell with an optical path of 10 m and a liquid-nitrogen-cooled, small metal dewar to cool the laser. The sensor performs stably at 30 ppb over a range of 10°C to 35°C, has a sensitivity of 7 ppb RMS at a fixed temperature, and is 32 cm x 50 cm x 20 cm.

A large telescoped infrared sensor was successfully flown on a rocket. A supercritical helium cryogenic system was used to cool the various parts of the sensor and to avoid cooling problems associated with liquid/gas phase mixing during the weightless part of the flight. The design and operational experience are presented. Since the flight of the sen-sor lasted only eight minutes, most of the operational time was spent calibrating and testing the sensor on the ground. This testing was relatively expensive since approxi-mately 20,000 liters of liquid helium were consumed. Alternate approaches to the design and operations are discussed which can improve the economy of the calibration and testing operations.

Video cameras can be used for surface temperature estimation from the measured radiance at single wavelengths or from the ratio of radiance at two wavelengths. This paper discusses limitations of video thermography for both methods in the near-, mid-, and far-IR regions.

An experimental set-up has been built to determine thermophysical constants of samples of various materials at room temperature. The same experimental equipment is used to evaluate the joint resistance of a two- layer sample. Theorical calculations are in good agreement with experimental results.Under certain circumstances, thermal losses and flash lamp efficiency can be taken into account.

HgCdTe photovoltaic arrays are developed in 3-5 micron and 8-14 micron wavelength regions. The average detectivities of linear arrays are 3x1ecm.He/W and 2x1e am.Hz1VW, respectively. Good spectral response uniformity within different elements of an array is obtained. HgCdTe material is grown by the solid state recrystallization and traveling heater method with Te as the solvent. The device performance correlation with crystal quality is investigated. The I-V characteristics of HgCdTe photodiodes severely degrade due to precipitates. Active area expension of photovoltaic detector made on slices with high density substructure is obvious. The higher density of subgrain boundaries of the materials, the greater crosstalk of the arrays. Detector performance dependence on slice surface orientation is observed. It shows that n-p junction made on (111) plane is of preferable. Experiments show that the detectivity of photodiode has positive correlation with below-gap optical transmittance of slice. The optical transmittance can be used as one of selection criteria for materials.

The digital contour enhancement techniques of infrared image are discussed. Emphasis is laid the thermal spread compensation method. On the basis of describing the theory of the method, the model is suggested. The concrete project based on the model for realizing digital contour enhancement of the infrared thermal image is put forward, and some test results are shown.