This paper sets the scene for the session 'Infrared in Australia.' A summary of the history of IR research and development in Australia is given, mainly spanning the years 1970 to the present. The paper then outlines the main research thrusts in the IR field in Australia. These include materials work, measurement, IR propagation, modeling, scene generation, devices, image processing, and IR systems. In addition brief mention is made of sensor and data fusion.

Two aspects of work on automatic target detection and tracking for electro-optic (EO) surveillance are described. Firstly, a detection and tracking algorithm test-bed developed by DSTO and running on a PC under Windows NT is being used to assess candidate algorithms for unresolved and minimally resolved target detection. The structure of this test-bed is described and examples are given of its user interfaces and outputs. Secondly, a development by Australian industry under a Defence-funded contract, of a reconfigurable generic track processor (GTP) is outlined. The GTP will include reconfigurable image processing stages and target tracking algorithms. It will be used to demonstrate to the Australian Defence Force automatic detection and tracking capabilities, and to serve as a hardware base for real time algorithm refinement.

We describe the specifications and design of a 1 - 5 micrometer camera (IRC-UNSW) being built at the School of Physics, University of New South Wales, for use on the 4-m Anglo- Australian Telescope (AAT) and the 2.3-m telescope of the Australian National University. The camera design uses three off-axis mirrors, allowing it to correct the off-axis aberrations in the telescopes themselves, and to obtain images with FWHM blur circles of 10 micrometer over a wide field-of-view (4.5 by 4.5 arcmin on the AAT, 5.8 by 5.8 on the 2.3-m). The camera accepts a collimated beam and can be used with an external Fabry-Perot etalon. We have fully characterized the performance of the camera with respect to misalignment and manufacturing errors. A thermal model has allowed us to calculate cool-down times, cryogen usage, and distortion due to small temperature differences in the optical system.

The Antarctic plateau has the potential for being the best site on Earth for conducting astronomical observations from the near-infrared to the sub-millimeter. Particular gains are expected in the 1 to 5 micron region, where the high altitude, low water vapor content, and low thermal emission from the atmosphere combine to create observing conditions unequalled elsewhere on the surface of the earth. We describe an instrument, the infrared photometer- spectrometer (IRPS), that we are using to quantify site conditions at the South Pole by measuring the near-infrared sky brightness. We also describe some of the unique problems associated with building instruments to work in Antarctica.

The spatial and temporal noise factors that set the ultimate performance limits on dynamic infrared scene projectors are assessed. Fixed pattern noise and the needs for non-uniformity correction procedures are both discussed, followed by a detailed analysis of the temporal and residual spatial noise contributions within the projector/imaging unit-under-test (UUT) composite system. It is found that the system is more tolerant to projector spatial noise by a factor of about four compared to UUT spatial noise. The temporal and spatial noise contributions are combined appropriately in order to provide guidance as to the levels of array uniformity that will need to be achieved if the expected demands for fine temperature resolution are to be met.

Modern imaging systems -- including CCD cameras, and various types of thermal imagers -- are capable of producing images in which useful information covers a signal range which is too great for current display technology to present to the human observer. Additionally, conventional display of electro-optical imagery severely biases the range of small signal differences that are observable by a viewer, despite the partial compensation provided by the approximately quadratic response of cathode ray tube monitors. This paper proposes the use of dynamic range zooming -- analogous to spatial resolution zooming -- as an efficient technique to support visual search of high dynamic range imagery, with the zooming being perceptually unbiased. The technique can be implemented as a lookup table in the signal processing chain of an imaging system. The effect of quantization in the output of the look up table is considered, as such quantization may be a problem for implementation.

An infrared imaging system which uses a staring focal plane detector array is described. The system operates in the 8 micrometer to 10 micrometer wavelength region and is required to acquire images with large scene temperature ranges. A nonlinear compensation algorithm is used to achieve a low NETD. The modeling and development of the imaging system and compensation processor required testing and characterization of the detectors comprising the array. The techniques used for the characterization of the array and some aspects of the use of the array are described.

This paper presents an overview of the research program on uncooled IR sensitive bolometer technology at the Defence Science and Technology Organisation (DSTO). It covers the design, materials growth, and performance of semiconductor film bolometer (SFB) IR detectors, with a strong emphasis on the practical considerations involved in their production. The overview includes a discussion of thermal isolation and the effects that vacuum packaging, and other alternative technologies, have on the performance of the detectors. Some of the principal applications of uncooled SFB IR detectors are also discussed. Finally, current technological developments in the DSTO uncooled sensor program and the direction of future research are described.

The along track scanning radiometer (ATSR) is an instrument for the accurate measurement of sea surface temperature, originally developed by the Rutherford Appleton Laboratory (RAL) in the United Kingdom, with substantial design inputs from the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Australia. ATSR-1 was launched on the European ERS-1 satellite in 1991, and has delivered high quality data. Auspace undertook the redesign, fabrication, integration, and test of the infrared focal planes for the follow-on mission, ATSR-2. This was delivered to RAL for integration into the overall instrument, which was successfully launched in April 1995. The Australian role on the follow-on advanced along track scanning radiometer (AATSR) represents about 30% of the overall instrument program, including the redesign, fabrication, integration, and test of both the visible and infrared focal planes for AATSR. The AATSR is scheduled to fly on ESA's ENVISAT program in 1999.

The performance of Hg_1-xCd_xTe long wavelength infrared (LWIR) photoconductors is strongly dependent on the semiconductor surface conditions and contact characteristics. In this paper we review these effects in relation to obtaining an optimum device technology suitable for use in two-dimensional infrared focal plane arrays (IRFPAs) based on the fabrication of high performance LWIR photoconductors on epitaxially grown Hg_1-xCd_xTe. Although the proposed design can be applied to a variety of epitaxially grown Hg_1-xCd_xTe material, for optimum performance the starting Hg_1-xCd_xTe semiconductor consists of epitaxially grown heterostructure layers in which a two-dimensional mosaic of lateral design photoconductors are fabricated. The heterostructure layer provides high performance devices at greatly reduced power dissipation levels, while the unique design allows for the high density integration of photoconductors in a two-dimensional array geometry with high fill factor. The proposed photoconductor array with n⁺ blocking contacts has been experimentally verified in a 3 X 3 array format with all elements in the array exhibiting background limited infrared photodetector (BLIP) performance at 80 K. Performance issues such as response uniformity, pixel yield, fill factor, crosstalk, power dissipation, detector impedance, array architecture, and maximum array size are discussed in relation to the suitability of the proposed photoconductor structure for use in IRFPA modules. It is found that in many cases the proposed photoconductor technology has the potential to deliver significant advantages, such as higher yield, higher fill factor, better uniformity, less crosstalk, and larger potential array size, in comparison to an IRFPA design based on photovoltaic technology.

This paper reviews the development and present status of mid-wavelength infrared (MWIR) HgCdTe photodiode technology in Australia. MWIR n-on-p photodiodes have been produced by both mercury in-diffusion and by boron ion-implantation on Hg-vacancy doped p-type HgCdTe. The photodiodes are planar structures passivated with thermally evaporated ZnS. High performance MWIR photodiodes have been developed with R_oA product of 2 X 10⁶ (Omega) cm² at 77 K. Recent work has focused on developing novel passivation processes based on anodic sulphurdization in which thermally evaporated ZnS is subsequently anodized in a non-aqueous sodium sulphide solution. Preliminary results suggest that this improves the insulating properties of thermally evaporated ZnS layers. The leakage currents, as measured through a 1 mm² MIS structure at a 1 V bias at 77 K, for the anodized ZnS layer are in the sub-picoamp range for all devices fabricated, whereas for the non-anodized layer the leakage currents are typically much larger by up to 3 - 4 orders of magnitude.

This study is concerned with the temperature dependent noise and responsivity performance of n-type x equals 0.32 Hg_1-xCd_xTe photoconductors. The fundamental noise sources that ultimately limit the specific detectivity, D_(lambda )^*, at temperatures T > 80 K are identified and correlated with the experimental material parameters of the device. A device model is presented for the responsivity and noise voltage which takes into account surface effects such as surface recombination and accumulation layer shunting. The model is in good agreement with the experimental values over the full temperature range from 80 to 300 K. Using a combination of experimental results and model calculations, the optimum device thickness and free electron concentration are presented which maximize D_(lambda )^* for a given operating temperature. Furthermore, it is shown that under ideal conditions it is possible to achieve background limited performance at temperatures up to 210 K. Experimental results are presented for responsivity, noise voltage, semiconductor surface charge density and D_(lambda )^* for an n-type x equals 0.32 Hg_1-xCd_xTe photoconductive detector as a function of temperature in the range 80 to 300 K. For a signal wavelength of 4 micrometer and a 40 degree field of view, a background limited D_(lambda )^* of 3.8 X 10¹¹ cmHz^1/2W^-1 was obtained for temperatures up to 180 K, while D_(lambda )^* of 1.4 X 10¹¹ and 2 X 10⁹ cmHz^1/2W^-1 were measured at 200 K and 300 K, respectively.

In this paper, the laser beam induced current (LBIC) technique is applied to the study of HgCdTe infrared materials and devices. In particular, a new quantitative model that relates the LBIC signal to photodiode device parameters is fitted to experimental data obtained from a mid-wavelength infrared (MWIR) HgCdTe photodiode. In addition, the LBIC technique is used to map and identify electrically active defect structures resulting from a novel HgCdTe passivation process in which thermally deposited ZnS is subsequently anodized in a non- aqueous sodium sulphide solution.

The application of HgCdTe as an infrared sensor material requires an accurate knowledge of the electrical parameters which are usually obtained from measurements of both the Hall coefficient and electrical conductivity as a function of temperature and magnetic field. Interpretation of the experimental data is difficult due to the unique band structure of HgCdTe and due to the presence of defects, surface layers, influence of the substrate, as well as localized non-uniformities. All of these effects can lead to 'anomalous' magnetic field dependence of the Hall data. In this paper we present the state of the art analysis of magnetic- field dependent Hall data including the recently developed quantitative mobility spectrum analysis (QMSA). This technique requires no prior assumptions to be made as to carrier type or number, and gives quantitatively accurate results for each carrier concentration and mobility. Of particular interest is a comparative study of the results obtained from QMSA with those obtained from previously developed techniques such as the Beck and Anderson mobility spectrum (MS) analysis and the recently described hybrid approach proposed by Meyer et al. which combines MS analysis with a multi-carrier fitting (MCF) procedure (MS + MCF). This comparison is illustrated with the analysis being performed on experimental Hall measurements from an epitaxial HgCdTe film which displayed 'anomalous' Hall characteristics.

The performance enhancement that can be achieved by using wider bandgap Hg_1-xCd_xTe as a capping layer for long wavelength infrared Hg_1-xCd_xTe photoconductors has been studied using both theoretical and experimental results. In the theoretical section it is shown that in the presence of a suitable energy barrier between the Hg_1-xCd_xTe infrared absorbing layer and the overlaying passivation layer, the high surface recombination rate which is usually present at the semiconductor/passivant interface does not have a significant effect on device performance. The energy barrier, which repels photogenerated minority carries from the semiconductor surface, is introduced by employing a Hg_1-xCd_xTe wafer which consists of a wider bandgap layer grown on a long wavelength infrared absorbing layer. A device model is presented which allows the responsivity to be calculated by taking into account surface recombination at both the front and back interfaces, thickness of capping and absorbing layers, recombination at the heterointerface, and variations in equilibrium carrier concentration. Experimental results are presented for x equals 0.22 n-type Hg_1-xCd_xTe conventional single layer photoconductors, and for heterostructure photoconductors consisting of an absorbing layer of x equals 0.22 capped by an n-type layer of x equals 0.31. Using the derived model, a heterointerface recombination velocity of 250 cm/s is extracted for the two layer photoconductor. Experimental results indicate that photoconductors which include a wider bandgap capping layer give a responsivity improvement of more than five times that of a conventional single layer device. Furthermore, heterostructure photoconductors are shown to be insensitive to the condition of the semiconductor/passivant interface.

Methods currently used for the generation of wind roughened sea surfaces suffer shortcomings in that they either generate sea surface heights that are symmetric about the mean surface height, or that they generate wave slope fields that do not necessarily have the correct slope distribution or correlation. A computationally efficient solution to this problem was developed for the one-dimensional case, i.e. the up and down movements of a single point on a sea surface in time. The following paper describes a generalization of this approach in two dimensions, which allows generation of a two-dimensional slope field with both the correct distribution of instantaneous surface slopes and the correct correlation of these slopes. The problem of integrating the resulting correlated non-Gaussian slope field into a height field remains unsolved at this point in time, as the height fields obtained by a straight forward integration show some artefacts and are hence not entirely satisfactory. Nevertheless, the method is an improvement on some of the algorithms currently in use where a correlated slope field is generated by convolving a field of independent slope values, drawn from a Cox-Munk distribution, with a spatial filter derived from the Pierson-Moskowitz power spectral density (PSD) of surface heights.

A specialized transmissometer has been installed at a location exposed to open ocean near Perth, Western Australia. The pathlength is 3 km and is 1.5 m above high water. The objective is to conduct atmospheric transmission measurements within the marine surface layer in the 3 - 5 and 8 - 12 micrometer transmission windows to quantify the degradation in transmission as a function of sea and meteorological conditions. A further experiment is being undertaken to examine the statistical properties of rapid fluctuations in the intensity of the transmitted beams.

This paper starts with a review of the status of presently available IRST sensors and continues with the problems that they are facing, even after more than 25 years of development. The problems arise from uncertainty in target signatures, atmospheric effects, background clutter and different types of hardware and processing shortcomings. Yet IRST has a promising future thanks to increase in knowledge on target/background signatures, improved optics and electronics sensor fusion and algorithm development. Furthermore IRST has good perspective due to the increased number of applications for various platforms and scenarios. The perspective grows with the increased performance for lower price. It is important to consider here IRST as part of the platform system and to endeavor for close integration with other system components.

The United States Navy lags most countries in the deployment of infrared sensors for ship defense. The need for shipboard electro-optic (EO) sensors was formally documented in a Department of Navy mission need statement (MNS), approved on 3 June 1992 by the Chief of Naval Operations. In response to this documented need, the Assistant Secretary of the Navy (research, development and acquisition) [ASN (RD&A)] directed the conduct of a cost and operational effectiveness analysis (COEA). The COEA is the method used by the Department of Defense to analytically derive and assess capability alternatives that will satisfy the MNS. The COEA team was led by the Massachusetts Institute of Technology, Lincoln Laboratory and consisted of representatives from various military, technical, financial and business organizations. This paper describes the COEA process that was used to define and assess shipboard infrared and search and track (IRST) capability alternatives.

The horizon infrared surveillance sensor (HISS) project is a multi-year effort to investigate infrared search and track (IRST) technologies in the areas of single scan target detection, single infrared (IR) band target detection and IRST operation within a multi-sensor integration (MSI) context for surface Navy applications. Each phase of the HISS project investigates particular characteristics of IRST operation and provides a system with specific capabilities in order to support scheduled MSI experiments and test events. This paper introduces the HISS project and presents a summary of the system performance achieved during the most recent series of land based tests.

In 1982 SAT proposed for the first time a second generation detector in the design of FLIRs for the TRIGAT program, since then different types of IR equipment have been developed on the basis of this technology: (1) An infra-red seeker for the MICA missile. (2) Three types of IRST: VAMPIR MB for naval applications, SIRENE for the Army and OSF for the Rafale aircraft. (3) Three thermal imagers: Condor 1 for the mast mounted sight equipping the long range anti tank system, Tiger installed on the sight of the medium range antitank system, and Condor 2 for the pilot sight of the TRIGAT French-German helicopter. Infra-red detectors are MCT IR-CCD focal plane arrays developed by SOFRADIR with the objective of the best standardization possible in spite of different configurations and specifications for each program. In this paper, we intend to present the main features of this technology for these programs and the advantages obtained by comparison with the first generation in terms of performance. Industrialization of these products is starting now, and a specific effort has been made to standardize the components, especially the driving and read out electronics. A set of ASICs has been developed to make compact detection modules including a detector in his dewar, a cooling machine, and a proximity electronic.

The U.S. Navy Office of Naval Research (ONR) is developing electro-optical (EO) sensor testbeds based on second-generation infrared (IR) focal plane array (FPA) technology, to be used for research and system demonstration in both shipboard and airborne applications. These technology-base sensors will demonstrate the functionality as well as many of the operational characteristics necessary ultimately in fielded infrared search and track (IRST) systems. The shipboard application involves the augmentation of existing and developmental radar assets in countering the advanced, sea-skimming, anti-ship missile (ASM) threat for mast-mounted use aboard high-value surface combatants. The ONR shipboard IRST testbed, the focus of this paper, currently is completing development, integration, and testing activities; BDM Federal (Arlington, Virginia) has supported these activities under contract to ONR over the past three years. The shipboard IRST testbed employs two separate, scanning FPAs covering the long wavelength IR (LWIR) and medium wavelength IR (MWIR). Preliminary test results show both a good correlation to predicted performance and a high level of absolute performance in terms of sensitivity and stabilization.

In 1993 and 1994 TNO-FEL participated in the execution of experiments over the North Sea in order to derive the infrared atmospheric scintillation and beam deformation process. In 1993 the NATO MAPTIP trial allowed us to observe an infrared point source from a platform at sea with a 10 micron FLIR, at various ranges (0.5 - 5 NM) and in various directions. In 1994 an infrared source was used at fixed distances on a platform at sea, with heights at the platform ranging between 1.5 and 7 m above the sea level. Some additional sources, at different height levels, were installed later in the experiment. The receiver system was placed at 18 km on a pier near the Dutch coast. Recordings took place with the receiver at two heights, 40 and 15 m above the average sea level. A single recording consisted of 10 seconds of measurements at 25 Hz with a 64 by 64 Cincinnati IRC-64A camera in the 3 - 5 micrometer band. Also data in the 10 micrometer band is available. We have also executed experiments with a source on a boat sailing out from close by to the optical horizon. Examples of these recordings are presented. The data were analyzed for scintillation effects, atmospheric APSF effects, and refraction effects. These data are compared to the atmospheric conditions that were recorded simultaneously, in order to model the infrared scintillation effects with meteorological conditions. In the paper we describe the analysis and results on the atmospheric effects with meteorological condition and with range. Results of these comparisons are also shown. We discuss the effects that these results have on the performance of IRSTs, and how improvements can be made to the IRST based on these results.

Many U.S. Navy programs need to make predictions of ship-based thermal imaging sensor (TIS) and infrared search and track (IRST) performance against various threats. An important factor in calculating performance ranges is the atmospheric transmission. The energy from the target must first pass through the atmosphere before reaching the sensor and will be attenuated. The amount of attenuation depends on the sensor band and weather conditions and may have a considerable effect on the strength of the original signal. This paper describes a TIS modeling technique which allows simple, band-averaged estimates of atmospheric transmission to be made at any range and an IRST performance analysis tool, both of which make use of the NSWCDD weather databases for prediction of atmospheric transmission.

Infracal software allows captured thermal images to be accurately calibrated by using spectral information taken from the imaging system as well as known data on the object and its environment. A pair of internal or external, in-field reference sources are used for calibration. Calibrated images may be scaled and total in-band radiant intensity calculated in a precise waveband, for example, 3.0 - 5.0 micrometer. This allows fairer comparison to be made between measurements obtained from different imaging systems under different environmental conditions. Information used for calibration includes: detector spectral responsivity, scanner spectral transmission, telescope temperature and spectral transmission, reference temperatures and spectral emissivity, object spectral emissivity, background spectral radiance, atmosphere spectral transmission and radiance, linearity of image gray levels and radiance specification band. Infracal is particularly valuable when accurate remote measurements of temperature or radiance need to be made through a poor atmosphere or at long range. It also allows ordinary thermal imagers to be used as imaging radiometers and spectrally recalibrates images obtained from radiometers such as the Agema 900.

Methods for visualizing scalar and spectral measurements taken from multiple observation angles around an object are presented in this paper. Scalar data measured at varying orientation and elevation angles can be visualized in the form of a signature globe, and spectral data from different orientation angles as a ring chart. The new visualizations are contrasted with standard charting methods and are shown to supplement them considerably, presenting a clear picture of the overall thermal signature.

In this article, recent characterization measurements made with LIFTIRS, the Livermore imaging Fourier transform infrared spectrometer, are presented. A discussion is also presented of the relative merits of the various alternative designs for imaging spectrometers.

The advent of high performance infrared detector arrays together with recent advances in coarse grating fabrication now makes possible the design and fabrication of infrared spectrometers that cover the mid-IR atmospheric windows at high resolution with no moving components. For applications involving small sources at large distances this instrument approach can provide a substantial increase in sensitivity over Fourier transform spectrometers at the same resolution and spectral coverage. We describe the design evolution of the LLNL cross dispersive infrared spectrometer (CDIRS) remote chemical sensors operating in the two atmospheric windows from 2.3 microns to 4.2 microns. The spectral and mechanical performance of the first generation 1 meter focal length prototype echelle grating spectrometer (EGS) is presented. A second generation cryogenic (150 K) spectrometer (mini-EGS) utilizes a high dispersion silicon immersion grating to provide a compact design. This instrument will be flown in the winter of 1995 as part of the instrumentation suite for the DOE airborne multisensor pod system (AMPS) effluent research program.

We have developed stimulated dual-band infrared (IR) computed tomography as a tool to inspect the aging infrastructure. Our system has the potential to locate and quantify structural damage within airframes and bridge decks. Typically, dual-band IR detection methods improve the signal-to-noise ratio by a factor of ten, compared to single-band IR detection methods. We conducted a demonstration at Boeing using a uniform pulsed-heat source to stimulate IR images of hidden defects in the 727 fuselage. Our dual-band IR camera and image processing system produced temperature, thermal inertia, and cooling-rate maps. In combination, these maps characterized the defect site, size, depth, thickness, and type. We quantified the percent metal loss from corrosion above a threshold of 5%, with overall uncertainties of 3%. Also, we conducted a feasibility study of dual-band IR thermal imaging for bridge deck inspections. We determined the sites and relative concrete displacement of 12- in. and 4-in. deep delaminations from thin styrofoam implants in asphalt-covered concrete slabs. We demonstrated the value of dual-band IR computed tomography to quantify structural damage within flash-heated airframes and naturally heated bridge decks.

An overview is presented of diffractive optics and micro-optics technologies and their application to infrared imaging systems. The technology overview describes several fabrication methods, emphasizing the importance of diamond machining for infrared systems and materials, and discusses various optical functions commonly implemented with diffractive elements. Applications are discussed both as a general survey by citing an extensive list of references and as a more detailed presentation of three examples.

The wavelet transform has gained much in popularity recently. Although the concepts underlying the wavelet transform have been used for some time, it is only in the last seven years that it began to have an impact, especially on signal and image processing. Wavelets have applications in differential equations, signal processing, image and video compression, and many other domains. We provide a brief introduction to wavelets and wavelet analysis, and compare the wavelet and Fourier transforms. The wavelet transform allows us to analyze nonstationary signals, which the Fourier transform cannot. This is a very important property of wavelets. A wavelet decomposition makes it possible to analyze a signal both in time (or space) and frequency domains and is appropriate for multiresolution analysis. One interesting application of wavelets is image fusion. For this application we take the wavelet transform of images coming from different sensors (e.g., visible and infrared). This provides us with a multiresolution description of visible and infrared images. The two images are then merged at each level of resolution. Applying the inverse wavelet transform on the resulting image generates a new image which is a composite of the two original ones. This concept can be applied to more than two images whether they are in the same spectral band or not. Some results are presented and compared with the classical pyramidal algorithms of Burt and Toet.

Optical-quality chemical vapor deposited diamond has optical, thermal, and mechanical properties similar to those of natural Type IIa diamond, with the exception of low strength in synthetic diamond. Clear windows up to 60 cm in diameter and 0.75 mm thick have been produced. Domes with a diameter of 60 - 75 mm and a thickness of 1 mm have been grown and machined on the outer surface by a laser process that reduces the surface roughness to 1 micrometer. Major challenges for diamond windows and domes are: (1) increasing the mechanical strength, (2) improving the growth rate, and (3) improving the efficiency of polishing. Four methods of attaching thin protective coatings of diamond to infrared windows have been demonstrated. Other protective coatings developed recently include boron phosphide, gallium phosphide, zinc sulfide, diamond-like carbon, amorphic diamond and polymers. Sapphire is the most durable, commercially available infrared window material, but its high temperature performance is limited by inability to withstand compression along the optical axis of the crystal. The compressive strength loss from the 20 degree Celsius value is 84% at 400 degrees Celsius and 97% at 600 degrees Celsius. Pros and cons of using ZnS, ZnSe, GaAs, GaP, Ge and Si as 2-color (midwave and long wave) windows are discussed. Engineering alternatives must be considered for extending the range of operating conditions for infrared window and dome materials.

Measurements of the state of polarization of the output of a double-clad fiber laser with relatively large birefringence and dichroism as a function of pump power and externally induced anisotropic stress are reported. The laser field ellipticity and the laser gain dichroism show a sharp modulation across the optical spectrum; this spectral modulation has a characteristic length of 1.75 nm, determined by the fiber core birefringence. The laser output power is partially elliptically polarized with the degree of polarization having a value near zero below and well above laser threshold, and near one close to threshold.

The responsivity of large scale platinum silicide arrays, having small pixels, is low compared to the responsivity of large area test diodes fabricated on the same wafer. Often, the responsivity loss is described by assigning a lower Fowler emission coefficient to the detectors. We find the reduced responsivity to be the direct result of a reduction in the effective active area of the detector. This reduction in effective active area becomes more pronounced as the detector cell size is reduced. We provide a simple model for the area reduction in terms of modulation of detector Schottky potential by the underlying depletion region of the detector guard ring. We also suggest changes in the detector array unit cell design, which will maximize responsivity.

Field deployable, high frame rate camera systems have been developed to support the test and evaluation activities at the White Sands Missile Range. The infrared cameras employ a 640 by 480 format PtSi focal plane array (FPA). The visible cameras employ a 1024 by 1024 format backside illuminated CCD. The monolithic, MOS architecture of the PtSi FPA supports commandable frame rate, frame size, and integration time. The infrared cameras provide 3 - 5 micron thermal imaging in selectable modes from 30 Hz frame rate, 640 by 480 frame size, 33 ms integration time to 300 Hz frame rate, 133 by 142 frame size, 1 ms integration time. The infrared cameras employ a 500 mm, f/1.7 lens. Video outputs are 12-bit digital video and RS170 analog video with histogram-based contrast enhancement. The 1024 by 1024 format CCD has a 32-port, split-frame transfer architecture. The visible cameras exploit this architecture to provide selectable modes from 30 Hz frame rate, 1024 by 1024 frame size, 32 ms integration time to 300 Hz frame rate, 1024 by 1024 frame size (with 2:1 vertical binning), 0.5 ms integration time. The visible cameras employ a 500 mm, f/4 lens, with integration time controlled by an electro-optical shutter. Video outputs are RS170 analog video (512 by 480 pixels), and 12-bit digital video.

We developed a 256 by 256 element HgCdTe hybrid infrared focal plane array (IRFPA) for the 8 to 10 micrometer band. We used three technologies to develop this high-performance, long-wavelength, large-scale IRFPA. The first innovation was to glue a sapphire substrate to a thinned Si readout circuit to reduce the thermal expansion mismatch with a HgCdTe diode array fabricated on a CdZnTe substrate. The second was to fabricate an interlaced switched- FET readout circuit using a 3 micrometer CMOS process. This readout circuit has a storage capacity of more than 10⁷ electrons and two video outputs capable of a 3.5 MHz data rate. The third was a HgCdTe diode array with an anodic sulfide passivation film and an optimized cutoff wavelength to reduce dark current and achieve high sensitivity. The noise equivalent temperature difference (NETD) was 0.06 K using f/2.5 optics. After 1000 thermal cycles (300 K - 80 K), there were no significant indium bump failures nor notable degradation in detector performance.

The performance of long wavelength infrared (LWIR) n⁺-on-p HgCdTe photodiodes is re-examined theoretically. It is assumed that the performance of photodiodes is due to thermal generation governed by the auger mechanism. The investigations are carried out for photodiodes operated in temperature range between 300 and 50 K. The effect of doping profiles on the photodiode parameters (R_oA product, I-V characteristic, photoelectrical gain and noise) is solved by forward-condition steady-state analysis. The theoretical predictions of photodiode parameters are compared with experimental data obtained by the research group at the LETI. Excellent agreement between both types of results indicate that experimentally measured performance of n⁺-on-p HgCdTe LWIR photodiodes are determined by the auger thermal generation mechanism.

A demonstrational thermal imaging LWIR camera system is described which is based on photovoltaic Pb_1-xSn_xSe-on-Si infrared sensor arrays. Epitaxial Pb_1-xSn_xSe layers, about 2 micrometer thick, are grown by molecular beam epitaxy onto 3 inch Si(111) substrates, and employing an intermediate CaF₂ buffer layer of only 2 nm thickness for compatibility reasons. Linear arrays with 256 pixels on 50 micrometer centers are fabricated in the layers with a batch photolithographic technique. Cut- off wavelength is about 10 - 10.5 micrometer at operating temperatures of 80 - 120 K, and quantum efficiencies greater than 60%. The sensors operate near BLIP. The read-out electronics chips contain, for each sensor, an integrator with a low noise JFET input transistor, correlated multiple sampling, and a sample and hold amplifier. They are wire- bonded to the sensor array. The JFET input transistors allow it to amplify from much lower source impedances (down to less than 10 kOhm) than with CMOS design without adding significant noise. Infrared sensors with lower impedances, operation at higher temperature, or sensors with longer cut-off wavelengths can therefore be used.

A hybrid HgCdTe 256 by 256 FPA for LWIR detection was fabricated and an infrared image was demonstrated. MCT epilayers were grown on GaAs substrates by MBE and annealed to p- type. The n⁺ on p photodiodes were formed by boron ion implantation. The mean value of zero bias differential resistance for the diode array was measured to be 8.0 M(Omega) with a cutoff wavelength of 9.5 micrometer. The effective quantum efficiency was estimated to be 0.55, and the optical cross talk was estimated to be 8.2%. A multiline parallel integration readout circuit designed especially for this 256 by 256 LWIR FPA, had 8.3 X 10⁷ electron capacity, a 190 microsecond integration time, and a single output. This work shows that the MBE growth method on GaAs substrates, pn junction formation process, the MLPI circuit design, and the hybridization technique are useful technologies.

We have fabricated unstressed gallium-doped germanium (Ge:Ga) far-infrared photoconductors for astronomical observations in space. A photoconductor's size as small as 0.5 by 0.5 by 0.5 mm³ is designed to reduce the influence of cosmic high-energy particles. The concentration of Ga in the Ge crystal is 1 X 10¹⁴ cm^-3. The responsivity of this type of Ge:Ga photoconductor has been improved by decreasing the Ga concentration from 2 X 10¹⁴ cm^-3 to 1 X 10¹⁴ cm^-3. Extrinsic semiconductor photoconductors, such as Ge:Ga, which operate at low temperature (less than 4.2 K) and low background photon influx (less than 1 X 10⁸ photons/sec) for space infrared astronomy, have undesirable behavior such as spike noise, hook response, and slow transient response. These phenomena limit the usefulness of the device as a photodetector. The optimum operating temperature of our Ge:Ga photoconductors at which we get the best noise equivalent power (NEP) is 3 K. The response to step increase in influx has a large amplitude with a time constant within 1 sec. This transient response of the photoconductor can be fitted by the model with three time constants. The curve-fitting shows the existence of a component with a time constant of around 0.05 - 0.5 sec, which we define as a moderate-fast transient response. We derived the time constant and the amplitude fraction of the moderate-fast transient response as a function of bias fields at several temperatures. The amplitude ratio of this component to the total response decreases as the bias field increases, and the time constant shortens as the operating temperature rises. Following, we estimated hole life time at the low temperature and under the low background photon influx from hole mobility and responsivity. Moreover, we found good agreement between the time constant of the moderate-fast transient response to the transient time that Hiromoto et al. (1992) pointed from a two-region model of a Ge:Ga photoconductor with ion- planted ohmic contact.

We have constructed a near-infrared camera with a 1040 by 1040 PtSi CSD array for astronomical use. The camera is attached to the prime focus (f/3.1) of the 105 cm Schmidt telescope at Kiso observatory. The field of view is 18.4 by 18.4 arcmin² and the spatial resolution is 1.06 by 1.06 arcsec²/pixel. The camera can be used mainly in J ((lambda) _eff equals 1.25 micrometer), H (1.65 micrometer), and K' (2.15 micrometer) bands. Since thermal emissions from the atmosphere and room-temperature bodies are main background noise sources in the near-infrared, we designed a cold baffle inside the camera to minimize the effect of the thermal radiation from the telescope. Both the charge transfer efficiency and the dark current of PtSi arrays are sensitive to the array temperature. Therefore we carefully control the temperature at 60 +/- 0.05 K by using a refrigerator and a temperature controller. The readout noise was reduced to 70 e by adopting a correlated multiple sampling technique. The array response was linear within 0.7% accuracy below 25% of the full well capacity (< 4.0 X 10⁵ e).

Test stations that are designed for today's imaging focal plane testing requirements normally include A/D conversion. The accuracy of the A/D conversion relates to the measurement accuracy of the test station and as such is a very important component of the stations end-to- end performance. Therefore characterization of the A/D conversion component is essential to the understanding of the test station's capabilities. This paper discusses the performance of two common A/D conversion devices integrated for use in a typical test station environment; 12 bit, 10 MHz; and 16 bit, 2 MHz. It details the realities of the accuracy of these A/D components as they are integrated into test stations, comparing the theoretical limits to the tested performance. Also discussed are the problems of testing high resolution A/D components at Nyquist frequencies at full scale input signals. The effect of A/D performance upon other common components within the test station is shown. This paper explores why time domain testing of A/D components may be a better gauge of performance than the more common frequency domain testing for evaluation within the focal plane testing environment.

The infrared response of polycrystalline and epitaxial CoSi₂/Si Schottky diodes with similar silicide thickness has been measured. For the polycrystalline diodes the quantum efficiency is found to be two times higher than for the epitaxial diodes, although both types of diodes present very similar barrier height. The observed improvement is attributed to grain boundary scattering of the excited carriers.

We have developed a smart CCD image sensor/readout with on chip analog processors. CCD signal processors are used with the PtSi detectors on the FPA to process image data. This smart CCD image sensor/readout has the advantages of CCD technology, i.e. low noise floor, 80 - 100 dB dynamic range as well as image readout through an array of analog (column) processors to perform parallel image processing operations. Each processor can implement firmware programmable functions such as: bi-directional, multiple color and IR, time delay and integrate function (TDI), region of interest windowing, transversal (temporal) and spatial filtering, background subtraction, exposure control and thresholding. The developed sensor is suitable for RECCE, and laser radar applications.

We discuss the development and the performances of a very long wavelength (13.5 - 15 micrometer) 128 by 128 AlGaAs/GaAs multiquantum well infrared imaging system. A highly uniform, high-yield QWIP focal plane array was hybridized to a CMOS multiplexer operating in a direct injection mode. For efficient light coupling an integral random scattering reflector (random grating) was incorporated. Due to the high uniformity, excellent imagery, low noise as well as a noise equivalent temperature difference (NE(Delta) T) of less than 30 mK were obtained when operating around 45 K. Therefore, high image contrast signal to noise ratio has been achieved.

In 1986, the first model K506A micro IDCA (integral dewar cooler assembly) was introduced as a pioneer in the market of the infrared thermal imaging community. Since then, more than 2000 K506 coolers have been sold by Ricor and fielded in a variety of electro-optical instruments. A vast majority of these microcoolers have been integrated with Cincinnati Electronics detector/dewar assemblies. During the last year, two additional models of this family were introduced, model K508A -- having 500 mW at 80 K cooling power, and model K543 -- having 1000 mW at 80 K cooling power. These two models are in final qualification programs and are in a preliminary fielding process. A general view of these models is presented. As part of the qualification program, a life demonstration test is carried out under the supervision of EL-OP LTD. The mean time to first failure as based on the June 12, 1995 status of these ongoing tests is: MTTF > 3000 hours for the model K508A and MTTF > 3800 hours for model K543. Testing is continuing until all units have failed, as the coolers continue to run, the MTTF values will increase proportionately.

The influence of temporal and fixed pattern noise in staring and scanning IR systems on the perception of MRTD test patterns is investigated. The objective of the study was to validate the U.S. FLIR92 model. Digital simulation techniques were used to provide the imagery required for the perception trials. The experimental results are compared to FLIR92 model predictions. It was found that the perception of MRTD test patterns in fixed pattern noise is a random process, with the result depending on the location of the test pattern. Predictions can be made for the average (50% probability) perception capability only. With this restriction, and using a variable threshold signal-to-noise ratio, the FLIR92 predictions were validated for staring systems and fixed pattern noise. Large discrepancies were found for fixed pattern noise in scanning systems. The threshold signal-to-noise ratio function was extracted from measurements made with temporal noise.

An IR target simulating projector system may sometimes require zooming lenses to simulate continuously the target's increasing size as a missile closes the range. In addition, special opto-mechanical layouts may sometimes be needed to allow cascaded action of two or more zooms to increase the zooming range, or to present the missile simultaneously two or more spectral regions. The zooms in these systems may differ in some important characteristics from the more conventional types: (1) The maximum speed and acceleration of their moving parts, as required by the dynamics of the missile-target approach scenario, are to be limited by the performance of commercially available stages carrying those parts, and (2) the motion control system must assure strict adherence to the 'zoom function' (relative position of the zoom's moving parts) to avoid defocusing during the zooming phase. In this paper we describe the implications of complying with these requirements on implementations actually designed and executed.

We describe a new pilot tactical training aid recently delivered to selected U.S. Marine Corps helicopter squadrons. The trainer is directed at improving targeting performance on missions supported by cockpit display of FLIR imagery. Particular features of the target identification sensor performance (TISP) trainer reported here include infrared phenomenology training, ground target identification drill, sensor controls operation training, symbology familiarization and performance prediction. Training is accomplished using real infrared imagery modified to reflect varying environmental conditions. Trainees can select the tactical scene and any of nearly 200 environmental combinations. Performance prediction graphs are then presented along with the modified images, thereby illustrating the imagery at the predicted ranges. The infrared phenomenology training further demonstrates the environmental impacts.

Uncooled IR focal plane arrays employ thermal detection mechanisms including the resistive bolometric, pyroelectric, and thermoelectric effects. Readout electronics are incorporated at each pixel. Two thermal isolation structures are possible: hybrid and monolithic. The ultimate limit to the performance is set by temperature fluctuation noise, which depends upon the structure. The state-of-the-art of the approaches employing the three mechanisms is reviewed.

This paper discusses design concepts for staring focal plane arrays of thin film resistance bolometer IR detectors, integrated with advanced signal conditioning and readout microelectronic circuits on the same silicon chip. These monolithic detector arrays are being developed for smart infrared sensors operated at ambient temperature. It is intended that the on-chip VLSI microcircuit include both analogue-to-digital conversion and signal processing functions. A description is given of detector fabrication at the Defence Science and Technology Organisation (DSTO), Salisbury, Australia, and system performance is considered for detector arrays integrated with new CMOS signal conditioning and readout circuits under development at the National Defence Research Establishment and University of Linkoping, Sweden.

Large arrays of ferroelectric elements have considerable potential for thermal imaging applications, offering operation and performance which challenges the cooled semiconductor detectors. A hybrid array technology, exploiting ferroelectric ceramic materials is the basis of a successful range of two dimensional arrays. The success is based on hot pressed ceramic (operated in both pyroelectric and dielectric bolometer modes), and the technologies for solder bump bonding and element reticulation. Arrays have been made with up to 10⁵ elements at pitches down to 40 micrometer, thus providing high resolution in compact systems. Ferroelectrics have now been produced as thin films with high figures of merit and, once deposition techniques compatible with silicon integrated circuits have been demonstrated, these will allow a more direct array fabrication. These integrated array technologies have potential for high yield and low cost arrays of very large numbers of elements. Thermal imaging systems research has led to a signal conditioning architecture optimized for pyroelectric and dielectric bolometer arrays and exploiting the ac coupling of the ferroelectric response to the radiation. The resulting compact sensors give high image quality with low fixed pattern nose in reliable, low power formats.

Loral Infrared & Imaging Systems is developing low cost, high performance uncooled infrared imaging products for both military and commercial applications. These products are based on the microbolometer technology, a silicon micromachined sensor which combines the wafer level silicon processing with a device structure capable of yielding NETD performance of better than 40 mK. To achieve a low cost sensor, Loral is proceeding with an integrated approach to the design and manufacturing processes associated with each major element of the uncooled sensor: focal plane array, electronics, optics and housings. Loral's 327 by 245 focal plane array has a 46.25 micrometer pixel pitch and incorporates a CMOS readout integrated circuit (ROIC). The ROIC has been designed to greatly simplify the external electronics, and features a single output which can operate at both 60 Hz (NTSC) and 50 Hz (PAL) video rates. The sensor electronics have been designed to meet a broad range of end user applications by providing both analog video and digital outputs with a large selection of user definable options and operating modes. To achieve low manufacturing costs across multiple end user applications, common optical interfaces, structural components, and manufacturing processes are being utilized. Sensor NETD is projected to be 40 mK normalized to f/1 and a 30 Hz frame rate. MRT is projected to be better than 0.1 degree Celsius at f₀.

Uncooled indium antimonide (InSb) photoconductors were studied to determine their primary noise source, spectral response, and maximum attainable spectral detectivity. The dominant noise component was determined by theoretically analyzing the background photon generation- recombination noise, thermal generation-recombination noise, and Johnson noise, using conditions of the experimental setup. It was shown that at room temperature (300 K) Johnson noise was predominant. Using this fact, an expression was derived for spectral detectivity which gave a value of 1.3 X 10⁸ cm-Hz^1/2/watt for a 25.2 k(Omega) photoconductor and 5.9 X 10⁸ cm-Hz^1/2/watt for a 10 k(Omega) photoconductor. The measurements from the experiment showed a peak wavelength response at 5 micrometer and a cutoff wavelength (50% of maximum) at 5.5 micrometer. The average peak spectral detectivity for the 25.2 k(Omega) photoconductor was 1.17 X 10⁸ cm- Hz^1/2/watt and for the 10 k(Omega) photoconductor, 4.48 X 10⁸ cm- Hz^1/2/watt. These results compared well with theory.

Solid-state micromachining techniques have been used in the fabrication of low-stress silicon nitride thermal isolation structures for pyroelectric detectors. Thin films of PbTiO(subscript 3/ and other related Perovskite materials are integrated on these structures. Process compatibility with NMOS electronics is demonstrated.

This paper presents the key design features of an uncooled infrared image sensor with 160 by 120 pixels. This sensor has a monolithic structure using micromachining technology. These features concern the configuration of the readout circuit, the structure of the infrared detector, and the thermal isolation structure in a pixel. The first feature is a simple readout circuit that includes neither an amplifier nor a switching transistor in the pixel. The second feature is the use of a thin film resistive bolometer made of polysilicon as the infrared detector. The detector has a P⁺-P^--N⁺ diode structure which operates as a bolometer and cuts off current passes through non-selected pixels. The forward resistance of the diode can be tailored by adjusting the shape and impurity concentration of the P^- region. Finally, a microbridge structure for the thermal isolation is made in each pixel by using the micromachining technology. The bolometer is monolithically integrated on this structure. Since polysilicon is generally used in the conventional Si-LSI process, this choice of detector material makes it possible to manufacture the image sensor using only current Si-LSI facilities, and realize a low cost uncooled infrared camera.

Choppers are inevitable parts of thermal imager or radiometer systems based upon uncooled pyroelectric focal plane arrays, since pyroelectric detectors require the incident radiation to be modulated. In comparison with other chopper techniques such as oscillator or shutter choppers rotating circular disks are considered superior in both mechanical simplicity and electronic control. However, the use of rotating disks introduces two non-ideal circumstances affecting the system's operation quality. Two-dimensional arrays require a bended chopper edge to approximate an ideal straight edge traveling over the array. Since sampling is performed line- sequentially, a local shift in time occurs between the sampling instant and the change of the optical chopper phase within each line. Furthermore, the chopper can only be placed in front of the entrance window of a detector array and, therefore, modulates the incident radiation in the blurred region of the optical path. This flattens the ideal rectangular modulation shape one would obtain if the chopper were able to run in the focal plane. Both effects attenuate the sensor output signal. The article presents a model for designing optimum chopper disks for 2- D arrays in order to achieve minimal possible signal attenuation. The optimization procedure incorporates all interdependencies between chopper layout, array size, chopper position, and readout velocity. The model also considers the unavoidable distance of the chopper plane from the focal plane and provides a chopper performance prediction method where the varying signal attenuation is discussed as contribution to the array's nonuniformity. Finally, a sample design is given, and measurements are presented.

A new pyroelectric linear array IR sensor has been developed by using the c-axis oriented La- modified PbTiO₃ (PLT:Pb_1-xLa_xTi_1-x/4O₃) thin films. The PLT thin films were prepared on (100) MgO single crystal substrates by rf-magnetron sputtering method. Without a poling treatment, this PLT thin film has high pyroelectric properties (x equals 0.1; PL10, (epsilon) _r equals 200, pyroelectric coefficient: (gamma) equals 5.5 X 10^-8 C/cm²K). For minimizing the substantial heat capacity, these IR sensors have a unique structure in which eight elements of pyroelectric thin films are supported only by polyimide resin films (thickness equals 3 micrometer). This sensor was fabricated by etching MgO substrates off thoroughly. These IR sensors have a high specific detectivity D^*(500,30,1) of 3.5 X 10⁸ cmHz^1/2/W. A new compact IR sensing system using this linear array sensor can measure thermal distribution (8 X 64) by horizontal sweeping of the vertical linear array elements. The image processing with neural network concept makes possible high-accuracy using a few data from the sensor elements. This sensor system is applied to a new type of room air conditioners to improve the comfort level control.

This paper analyzes the temperature rise during the pulse bias period of a resistance bolometer used for thermal radiation detection, and the effectiveness of two proposed methods for compensation of the output ramp voltage from the detector due to this temperature increase. The pulse bias heating is usually much larger than the incident radiation heating to be detected. In a focal plane array (FPA) of detectors, this heating may cause a voltage ramp equal to or larger than the fixed pattern noise (FPN) voltage caused by the pixel-to-pixel detector resistance tolerance. If no compensation is made, a large portion of the dynamic range in the succeeding redout circuitry must be reserved for this voltage ramp. This results in decreased temperature resolution of the sensor or increased complexity in the designed readout circuitry needed to meet specified performance. General expressions for the detector heating are derived to give an understanding of the important parameters. Also, two different compensation techniques are presented together with simulation results showing an effective compensation of the pulse bias heating over a 50 K ambient temperature range. An efficient compensation must consider pulse bias heating dependence on the bolometer resistance, which in turn is dependent on the chip temperature. Therefore, a sensor that operates at ambient temperature will require a compensation that considers both the chip temperature and the resistance tolerance from wafer to wafer. This is achieved by means of an on-chip sensing of the detector resistance to automatically adjust the amount of compensation.

The IR SnapShot^TM camera is an uncooled, handheld imaging radiometer designed primarily for the predictive and preventive maintenance market. It employs a linear thermoelectric array to scan the focal plane of a germanium lens, thereby producing a 120 by 120 pixel image with an instantaneous field of view (IFOV) of 2.8 by 2.8 mrad, an overall field of view (FOV) of 20 degrees by 20 degrees, a frame time of 1.44 sec, a noise equivalent temperature difference (NETD) no greater than 0.2 C, and a radiometrically calibrated dynamic range up to 500 C.

For several decades, the main thrust in infrared technology developments in Australia has been in two main sensor technologies: uncooled silicon chip printed bolometric sensors pioneered by DSTO's Kevin Liddiard, and precision engineered high quality Cadmium Mercury Telluride developed at DSTO under the guidance of Dr. Richard Hartley. In late 1993 a low cost infrared imaging device was developed at DSTO as a sensor for guided missiles. The combination of these three innovations made up a unique package that enabled Australian industry to break through the barriers of commercializing infrared technology. The privately owned company, R.J. Optronics Pty Ltd undertook the process of re-engineering a selection of these DSTO developments to be applicable to a wide range of infrared products. The first project was a novel infrared imager based on a Palmer scan (translated circle) mechanism. This device applies a spinning wedge and a single detector, it uses a video processor to convert the image into a standard rectangular format. Originally developed as an imaging seeker for a stand-off weapon, it is producing such high quality images at such a low cost that it is now also being adapted for a wide variety of other military and commercial applications. A technique for electronically stabilizing it has been developed which uses the inertial signals from co-mounted sensors to compensate for platform motions. This enables it to meet the requirements of aircraft, marine vessels and masthead sight applications without the use of gimbals. After tests on a three-axis motion table, several system configurations have now been successfully operated on a number of lightweight platforms, including a Cessna 172 and the Australian made Seabird Seeker aircraft.

The purpose of the work described in this paper is to study experimentally the fuel droplet temperature behavior in order to better understand the processes of the fuel injected in combustion chambers of engines and to validate theoretical models of droplets vaporization. This paper describes the temperature measurement of small fuel droplets using an IR detection system, composed of two coupled off-axis parabolic mirrors and a MCT LWIR detector. The system measures the temperature behavior of a chain of droplets with equal diameter and spacing. Our system has been used to quantify the temperature drop associated with 200 micrometer diameter droplets moving at a velocity of greater than 5 m/s in ambient air. Finally, we describe a feasibility study of how to do this measurement when the droplets are inside a flame. Initial experimental results are presented.

For many terrestrial IR imaging applications, speed and sensitivity requirements can be met by several sensor technologies, including PtSi staring arrays, second generation scanned MCT arrays, InSb, and MCT staring arrays. However, for high quantum efficiency staring arrays, MCT and InSb in particular, much of the available signal remains unused because of well-fill restrictions. In narrow spectral band imaging applications the available photon flux is dramatically lower and high quantum efficiency becomes critical. One such application is detection and imaging of gases. High quantum efficiency arrays allow the use of narrow band filters centered on the absorption band of the gas, thus enhancing the ability of the imager to discriminate between background and gas emission. Results of imaging studies performed on a camera with a narrow band filter for gas detection are presented and analyzed.

A high performance infrared intrusion sensor employing uncooled thin film resistance bolometer detector technology has been developed at the Defence Science and Technology Organisation. The sensor incorporates lightweight metal aspheric optics and novel digital signal processing to give a high detection probability at low false alarm rate. The equipment is robust, has excellent reliability and long life in field applications. It can operate in local attended mode or remotely with a dedicated communication link and network system, and can be integrated with existing unattended ground sensor (UGS) systems. The sensor is highly versatile, having applications in military remote surveillance, vital asset protection, and high grade security systems. Its performance has been demonstrated in a wide range of environments, from northern European winter to tropical Australian summer.

We demonstrate the use of an infrared focal plane array (IR-FPA) to measure the spatially resolved surface temperature of a GaAs MESFET (gallium arsenide metal-semiconductor field-effect transistor) under dc and rf operating conditions. By compensating for variations of the small emissivity, absolute temperatures of plus or minus 5 degrees Celsius, and small temperature differences of 1 degree Celsius can be determined. By deconvolution of the lens MTF (modulation transfer function) we attain a resolution of 6.25 micrometer. The combination of thermal and spatial sensitivity makes our set-up ideally suited for non- destructive characterization of semiconductor devices.

As FT-IR remote sensing techniques become more accessible, there are increasing interests to apply this open-path measurement method to detect and measure airborne pollutants. Thus a research for VOCs emission pollutants in the petrochemical industry park is conducted. In this study, we focused on the identification of the gaseous pollutants as well as the location of the VOCs pollutants from different factories. Measurement is sampled at every half hour period to obtain the time series plots of observed gas concentration for the gaseous pollutants. Besides the inherent components in ambient air such as carbon monoxide, methane, and ozone, the results of the measurement indicate that the major pollutants detected in this industrial park include vinyl chloride, chloroform, hydrogen chloride, 1,2-dichloroethane, 1,3-butadiene, ethylene, propylene, n-hexane, acetic acid, methyl acetate and ammonia. Some of these toxic pollutants are carcinogens and also the chloride related compounds are potentially a threat to the depletion of ozone. All of these measurements indicate that the pattern of the pollutants for each location is significantly different from each other pattern. In addition, the concentrations and the presence of absence of pollutants were dramatically affected by wind directions. Under this case, suspicious polluting plants are successfully being identified by examining the pattern of compounds, pollutant's concentration time series, metrology, and manufacturing process.

France has been involved from the very beginning in the development of infrared technologies. For fifty years now, French scientists and engineers have been working on the fascinating albeit sometimes elusive technologies which are required to master optronics equipment and at the heart of them, infrared detectors. A general overview of this adventure is presented starting with the activities in the late 1940s of the first company in France dedicated to infrared equipment, les Etablissements Jean Turck, to the present activities on second generation IRFPAs handled by Lir and Sofradir. Special attention is given to the scientific activities within the French Universities and the CNRS (Centre National de la Recherche Scientifique) as well as the industrial activities within first of all Sat, then Thomson CSF, and now Sofradir. This paper does not pretend to cover exhaustively all the IR detector activities of the last thirty years in France. Rather than trying to cover all these activities, which anyhow would be impossible, I prefer to focus on what, I think, have been the unique aspects of IR detector development in France. One is a technological issue, the PV choice focused on MCT. And the second is an organizational issue, the Lir-Sofradir mode of operation and cooperation.

Since 1985, when THOMSON-CSF/TCS became involved in detector development for remote earth imaging applications, the InGaAs technology has been greatly improved leading to spreading use in a wide range of domains. Space born applications in the SPOT 4, Vegetation (SPOT 4 host), IRS C/D, and Huygens/Cassini programs are among the most brilliant successes. Initial industrial demonstrations, which have come last, are contributing to InGaAs demystification. We present InGaAs main forces and difficulties, and how recent progress has transformed the product offer toward these main points. Figures concerning dark current, noise (especially burst noise), spectral response, non linearity, and modulation transfer function (MTF), which are of main importance in detector arrays are presented. Trade-offs are pointed out with respect to end application. InGaAs supremacy in the 0.9 micrometer to 1.7 micrometer spectral range is discussed -- dark current density down to 2E - 08 A/cm² and more than 85% quantum efficiency are currently obtained nowadays. Finally, InGaAs industrial maturity is emphasized as well as the strong and lively activity at THOMSON/TCS on technology which lets everyone expect increasing performances and recessing prices.

P(VF₂-TrFE) pyroelectric copolymer is chosen for its high level of compatibility with existing microelectronics processes, and convenient electrical properties for infrared (IR) 8 to 14 micrometer imagery in the performance range of NETD 0.1 K to 1 K. Low cost sensors, achievable thanks to the monolithic silicon wafer processing, standard package assembly, and uncooled operations, allow us to address a lot of low-end applications in which conventional IR imaging techniques -- high priced hybrid mercury cadmium telluride or indium antimonide arrays, liquid nitrogen cooling and sophisticated image processing -- are nowadays incompatible with large volume user's needs and market prices. The paper describes pyroelectric device trade-offs, architecture, and process. Based on the interline architecture, the sensor performances of the TH 7441A 128 by 128 area array infrared detector are presented: compatible with 1 inch optics, the square array is made of an 80 by 80 micrometer squared pixel, on a pixel pitch of 85 micrometer. The CCD multiplexer using patented on-site processing is designed to deliver the image information at the maximum rate of 50 image/s. Lower rates are achievable. Imaging performances are the following: a NETD of 1.7 K is achieved with an integration time of 10 ms and the use of a f/1 optics presenting a transmission of 0.8. Recent improvements in the properties of the pyroelectric sandwich include thermal insulation of the pyroelectric layer through mixed air-polyimide material and pixel side to side insulation. Thanks to on-wafer pixel reticulation, an increased modulation transfer function of 51% at Nyquist frequency is achieved.

Quantum well infrared photodetectors (QWIPs) form a new generation of infrared detectors based on carrier confinement in ultrathin semiconductor heterostructures. The artificial energy levels in these wells can be tailored to match any optical transition in the 3 - 20 micrometer photon wavelength range by adjusting the quantum well width and the barrier composition. In this communication, we summarize our present understanding of the physics of QWIP detection: photoexcited carrier emission and capture probability, contact injection, and noise mechanisms. We also present the performances of optimized devices for the infrared Detection in the 3 - 5 micrometer and 8 - 12 micrometer wavelength ranges. We also illustrate the major advantages of this new technology for infrared staring arrays: (1) standard III-V substrates and technology, thermal stability, uniformity, large areas, low development costs, radiation hardness, (2) adjustability from 3 to 20 micrometer, (3) new functions: multispectrality, spectrophotometry, band switching, optical reading.

Thanks to the continuing effort in the field of infrared detectors in France, Sofradir has now a large number of different IRFPAs operating in both wavelengths MW and LW. During the last two years, as the maturity of the technology was demonstrated and the producibility of complex components was achieved, Sofradir has focused recent activity on new products. The target of this paper is to focus on new products such as 128 by 128 LW IRFPA and 480 by 4/576 by 4 monolithic MCT IRFPA. Typical results are presented.

A 256 by 256 IRCMOS array with a 35 micrometer pitch operating in the (3 - 5) micrometer wavelength range at 77 K has been developed at LETI/LIR. The readout circuit was designed and processed with a 1.2 micrometer design rules standard CMOS technology. Photovoltaic (PV) detectors, made on liquid phase epitaxy (LPE) layers, are interconnected by indium bumps on the readout circuit. A description of the readout circuit is given and the main electro-optical characteristics of the component are presented. The detector arrays exhibit a quantum efficiency of 60% with a non-uniformity of less than 4%. The pixel operability is greater than 99.5 %. In spite of the non-optimized threshold voltage of the technology, the charge handling capacity reaches 20.10⁶ e^- for the 35 micrometer pitch. This allows an NEDT of 12 mK at the output midpoint. Excellent imagery has been obtained with this component operating at 77 K using f/2 optics. The imagery has high contrast with no significant fixed pattern noise.

New gallium-doped silicon 128 by 192 element arrays have been achieved at CEA-LETI-LIR (Infrared Laboratory) for imaging in the 8 - 14 micrometer spectral range. This program is in keeping with the previous detector development for the ISOCAM camera (32 by 32 element arrays) and for ground-based observation (64 by 64 element arrays). The main features of the new detectors are: a pitch of 75 micrometer which leads to 10 by 15 mm² chip dimensions, two selectable storage capacitors (respectively 0.1 and 0.5 pF), a DVR readout circuit achieved in an NMOS silicon line with 1.5 micrometer design rules. The main electro- optical performances are the following: a peak responsivity of 4.0 A/W, a noise of 58 fA rms over the 0.1 - 128 Hz spectral range which is very close to the BLIP noise, and a corresponding noise equivalent power of 1.4 10^-14 W.

The purpose of this paper is to present the Cryotechnologies SA research and development in miniature cryocoolers covering the following subjects: pulse tube miniature cryocoolers, improvement linear Stirling cryocooler reliability, IDCA with rotary and linear Stirling cryocoolers, gas mixture for an active joule Thomson system, and a new demand flow mechanism for joule Thomson cryostats.

In France, the real start of work on the applications of infrared radiations occurred around 1947 - 1948. During many years, technological research was performed in the field of detectors, optical material, modulation techniques, and a lot of measurements were made in order to acquire a better knowledge of the propagation medium and radiation of IR sources, namely those of jet engines. The birth of industrial infrared activities in France started with the Franco-German missile guidance programs: Milan, HOT, Roland and the French air to air missile seeker programs: R530, MAGIC. At these early stages of IR technologies development, it was a great technical adventure for both the governmental agencies and industry to develop: detector technology with PbS and InSb, detector cooling for 3 - 5 micrometer wavelength range, optical material transparent in the infrared, opto mechanical design, signal processing and related electronic technologies. Etablissement Jean Turck and SAT were the pioneers associated with Aerospatiale, Matra and under contracts from the French Ministry of Defence (DGA). In the 60s, the need arose to enhance night vision capability of equipment in service with the French Army. TRT was chosen by DGA to develop the first thermal imagers: LUTHER 1, 2, and 3 with an increasing number of detectors and image frequency rate. This period was also the era in which the SAT detector made rapid advance. After basic work done in the CNRS and with the support of DGA, SAT became the world leader of MCT photovoltaic detector working in the 8 to 12 micron waveband. From 1979, TRT and SAT were given the responsibility for the joint development and production of the first generation French thermal imaging modular system so-called SMT. Now, THOMSON TTD Optronique takes over the opto-electronics activities of TRT. Laser based systems were also studied for military application using YAG type laser and CO₂ laser: Laboratoire de Marcousis, CILAS, THOMSON CSF and SAT have developed during the 70s prototypes for a laser range finder, lidar, laser weapon, and target designator. The constant need to develop increasingly efficient infrared equipment led to a significant increase in the number of detector elements implying the integration of the detector and multiplexer electronic. After tests on several possible technologies at SAT, THOMSON CSF, and LETI, the work performed by these teams in 1980 was concentrated on the development of an MCT type IRCCD detector. The selection of this detector technology for the TRIGAT program led to the creation in 1986 of SOFRADIR with the pooling of the different existing expertise. Much other equipment of the first generation was created during the 80s and is now in production: IRST for naval and airborne applications; IR line scanner for airborne reconnaissance; light thermal imagers for man-portable weapons; infrared seekers for ground to air and air to air missiles; thermal sights for submarine, tank, and missile launch systems; night vision systems for flying helicopter and aircraft; air to ground attack pods for night and day operations.

In recent years, CEDIP has designed and produced several imaging systems using MCT-FPA from SOFRADIR. From the simple 64 element linear array to the advanced 128 by 128 LWIR matrix, a large amount of expertise has been gained in the design of advanced electronic modules and optics. The purpose of this paper is to present some of the technical solutions and basic concepts defined and used by CEDIP's engineers during the design phase. Performance data about the commercially available Peltier cooled IRFPA camera, based on these design concepts, also is given.

In this paper on-focal plane array function trends are reviewed regarding the CCD and CMOS silicon technologies. The analysis is carried out according to a classification of the IRFPAs into three categories following the integration density, array sizes, and pixel pitches. Then a brief presentation of SOFRADIR and THOMSON TTD OPTRONIQUE's cooperation to design and develop a new advanced IR system is given.

SPOT 4 satellite will carry two new observation instruments with infrared extended capabilities compared to the existing SPOT series (HRV). High spatial resolution imaging (20 m) will receive an additional 1.7 micrometer channel as well as a new payload instrument called VEGETATION. This instrument is devoted to low resolution (1.2 km) but wide field of view (2000 km) monitoring. Both IR channels use InGaAs THOMSON-CSF technology but each one has some specific features. Focal plane geometry, photodiodes configuration (staggered or lined up), dark current and fixed pattern noise reduction are discussed. Moderate cooling and thermal regulation around 5 degrees Celsius is achieved in order to meet the specified radiometric performances. SPOT's HRVIR and VEGETATION missions are reviewed from SPOT 1 to SPOT 4 and future SPOT 5 SWIR channel is introduced.

Satellite remote sensing measurements in the thermal IR spectrum have existed since the beginning of the space era with the first Tiros satellites in the sixties. The low spatial resolution data provided up to now have been mainly used for mesoscale applications in meteorology and oceanography. To study and understand on a local scale those natural phenomena which can be characterized by their radiative temperatures, the users need high spatial resolution data in multi-spectral channels with a good radiometric sensitivity. Thanks to recent progress in IR detector technology, these specifications will soon be satisfied. The IRSUTE (infra red satellite unit for terrestrial environment) instrument is designed to provide high spatial (50 m), multi-spectral (4 channels in the IR), high repetivity data with a good radiometric sensitivity. Moreover, the IRSUTE payload has to be compatible with the limited resources (mass, power, volume, ...) provided by a small satellite. The design of the instrument is based on the push-broom technique which uses IR-CCD linear array detectors positioned in the focal plane of a collecting telescope. The first results of a pre-feasibility study on the inboard instrument are presented.

In the frame of technological preparation for scientific Earth observation, the European Space Agency (ESA) has undertaken a study program for an infrared instrument intended to be implemented on a polar platform type satellite: HRTIR (high resolution thermal infrared radiometer) will be a support for the study of land surface processes by providing reliable information on both temperature and emissivity. The HRTIR instrument, designed under Matra Marconi Space (MMS) leadership, will observe the Earth in three IR spectral bands ranging from 8 micrometer to 12.5 micrometer with a spatial resolution of 50 m and a swath width of 50 km. One of the major challenges is to develop and implement 1000 elements linear arrays at such long wavelengths. SOFRADIR, with LIR support for technology, is responsible for the detectors development. A detector breadboard has been studied and developed to be operated in the most critical band, i.e. 11.5 - 12.5 micrometer. The breadboard is made of three HgCdTe modules mechanically butted and hybridized to two CCD multiplexers. To achieve the required 0.1 K NEdT, the detector is operated at 50 K; the mechanical butting is designed with the objective of zero dead pixel at the butting joints. The detector breadboard has been integrated in a focal plane and detection chain breadboard at MMS to be fully characterized. The results have proven the very high quality of LIR/SOFRADIR HgCdTe material beyond 12.5 micrometer together with a high photodiode production yield. The feasibility of mechanical butting was also demonstrated although further investigations and tests still have to be carried out to significantly increase the butting yield.

The TH74KB41A camera is designed for the THOMSON-CSF Semiconducteurs Specifiques (TCS) TH7441A pyroelectric IR sensor, optimized for the wavelength range 8 to 14 micrometers. This monolithic area sensor is made of 128 by 128 pixels (85 micrometers by 85 micrometers) compatible with a one inch diameter optic, and operates at ambient temperature. The camera is optimized to operate with a short integration period -- 10 ms -- allowing 50 frames per second, then compatible with a wide range of applications, including real-time imaging. Three video outputs are available: one 10-bit unprocessed image output every 10 ms, one 8-bit processed image every 20 ms, and one composite analog video signal using a 128 by 128 zone of the display. Eight-bit and analog outputs are obtained by summing the two successive fields -- alternately positive and negative. The 10-bit output delivers successively these positive and negative fields, 10 ms each. It is possible to select an analog gain 3 or 6 and a digital gain of 1, half, quarter or an eighth. Offset may be adjusted to fit the temperature range to be viewed in accordance with the chosen gain. The camera also produces pixel, line, and frame pulses to ease synchronization of an optional frame grabber. Power consumption is as low as 12.5 W, including chopper power supply. The useful scene dynamic range is plus or minus 170 degrees around the ambient temperature. With an f/1 optic -- optics transmission is 0.8 -- the NETD is around 1.5 K and the recognition 0.75 K. The contrast transfer function at Nyquist frequency -- 5.9 line pairs per mm -- is 38%. The small size of this lightweight camera body (200 mm by 125 mm by 60 mm without lenses) together with its low power consumption allow easy use everywhere. This camera benefits from the pyroelectric area sensor and does not use any cooling. This camera, quite cheaper than a camera using a cooled detector, is well suited to a wide range of applications such as laser beam profile analysis, IR machine vision in the industrial field -- useful for process control, security, night vision, etc., -- fire fighting or prevention, search and rescue in poor vision environment, medical diagnosis, etc.

SYNERGI is a set of modules developed in collaboration between Thomson TTD Optronique, Pilkington Optronics, and Carl Zeiss. This module set is dedicated to new generation multipurpose high performance thermal cameras based on the SOFRADIR 288 by 4 element IRCCD detector. The manufactured products, as well as the various camera configurations designed, exhibit the key features of SYNERGI: cost effectiveness, compactness, flexibility, reduced power consumption. Special care has been taken in the technical design to minimize noise sources (early video digitization, power supplies synchronization), to maximize boresight stability (specific scan sensor), to enhance image characteristics (zoom modes, digital outputs, external synchronization), to make easy maintenance (dedicated serial link, BITE).

Iris is a modular infrared thermal image developed by SAGEM since 1988, based on a 288 by 4 IRCCD detector. The first section of the presentation gives a description of the different modules of the IRIS thermal imager and their evolution in recent years. The second section covers the description of the major evolution, namely the integrated detector cooler assembly (IDCA), using a SOFRADIR 288 by 4 detector and a SAGEM microcooler, now integrated in the IRIS thermal imagers. The third section gives the description of two functions integrated in the IRIS thermal imager: (1) image enhancement, using a digital convolution filter, and (2) automatic hot points detection and tracking, offering an assistance to surveillance and automatic detection. The last section presents several programs for navy, air forces, and land applications for which IRIS has already been selected and achieved.

Typical targets to be detected in ground surveillance scenarios have two common properties that are physically distinct, motion and radiation. This applies to pedestrians, vehicles, trucks and light aircraft. Infrared sensors can be used for the detection of radiating bodies through the application of image processing techniques and Doppler radar is an excellent detector of radial body motion. RAPTOR is a system whose aim is to produce a ground surveillance device with higher detection performance and lower false and nuisance alarm rates than possible with either or both sensor types operating alone. RAPTOR combines two complementary sensors, an infrared imager and a pulsed Doppler radar, which automatically detect targets using different physical phenomena and then uses data fusion techniques to enhance the automatic target detection performance. The fusion process includes alignment, correlation, association, target data processing and multisensor management. RAPTOR uses the two sensors in parallel, bore sighted on a single rotational platform. Both sensors concurrently scan the spatial volume and the individual target detection results, 'soft-targets,' are combined using data fusion algorithms to generate a confirmed set of targets, 'hard-targets.'

In this paper, the constitutional characteristic and performance of an automated scattering measurement system is introduced. The angular distributions of laser scattering intensity for some rough plane samples are measured at 0.63 micrometer and 0.904 micrometer. The materials of these samples include aluminum, stell, carbon and paints. The dependence of angular intensities on dielectric constant, roughness, and wavelength are analyzed.

Conventional methods in robot vision use the intensity of light reflected or emitted by objects in order to perform object recognition. However, information contained in the polarization of the light can often aid in the determining of surface properties such as roughness, index of refraction, and spatial orientation. Imaging of such surface properties would facilitate image segmentation and classification of objects in military target recognition, environmental monitoring, oceanography, forestry, agriculture, and automated assembly. Physics Innovations Inc. is developing a thermal imaging technique where, in each image pixel, three Stokes parameters are sensed simultaneously and at video frequencies. The Stokes parameters are intensity I, percent of polarization P, and angle of the plane of polarization (phi) . Although infrared, thermal intensity images of terrestrial scenes have low contrast, images of P and (phi) are expected to have high contrast. In this paper the Physics Innovations sensor is described. We also discuss our evolution of the performance of a prototype sensor. Images of I, P, and (phi) from the prototype sensor demonstrate that, for common man-made objects with smooth surfaces, surface orientation can be derived. Surface orientations can be measured in the same image frame as temperature distribution. From our results using the prototype sensor, we conclude that three-dimensional information, in addition to thermal information, can be derived from polarization-sensitive, thermal imaging.

In 1994, Inframetrics introduced the InfraCAM^TM, a high resolution hand-held thermal imager. As the world's smallest, lightest and lowest power PtSi based infrared camera, the InfraCAM is ideal for a wide range of industrial, nondestructive testing, surveillance and scientific applications. In addition to numerous commercial applications, the light weight and low power consumption of the InfraCAM make it extremely valuable for adaptation to space borne applications. Consequently, the InfraCAM has been selected by NASA Lewis Research Center (LeRC) in Cleveland, Ohio for use as part of the DARTFire (diffusive and radiative transport in fires) space borne experiment. In this experiment, a solid fuel is ignited in a low gravity environment. The combustion period is recorded by both visible and infrared cameras. The infrared camera measures the emission from polymethyl methacrylate, (PMMA) and combustion products in six distinct narrow spectral bands. Four cameras successfully completed all qualification tests at Inframetrics and at NASA Lewis. They are presently being used for ground based testing in preparation for space flight in the fall of 1995.

Thermal imaging devices -- especially large two-dimensional focal plane arrays -- are useful for a wide rage of applications. A key factor that has limited thermal imaging mainly to military applications is the cost of the imaging system. The approach taken here to overcome the cost limitation involves the integration of a quantum well intersubband photodetector (QWIP) and a light emitting diode (LED) for making large two-dimensional focal plane arrays. The advantage of this unique new approach over the conventional one is that it uses a mature GaAs-based materials system and avoids hybrid bonding with Si multiplexer chips. A demonstration device has been made and is discussed here, with the QWIP operating at a peak detection wavelength of 9 micrometer and the LED emitting at 930 nm. The performance issues are also addressed.

Most existing forward looking infrared systems (FLIRs) are based on parallel scanning of the object by rotating a mirror across a vector detector which has a relatively small number of elements (typically 60 - 180). In order to get a good picture, it is necessary to fill all the lines of the display with information which has been achieved from a small number of elements. The benefits of this line filling (multiplication) operation which is an interpolation process is to increase overall system magnification hence improving the modulation transfer function (MTF) of the eye and the display. This process also allows the display of the image without geometric distortion on a pleasant, not too high, picture width to height ratio (aspect ratio). Concerning the MTF of the display, the line multiplication operates in two ways. It increases the MTF by scaling down the display Gaussian point spread function width due to increased magnification but, as we shall show, also decreases the MTF because after the line filling process, the image is composed of several sub-images each of which includes a different set of display lines. The MTF of each of these sub-images may be decreased and the sub-images themselves, their line of sight, may be shifted one to the other. In this work we shall quantify the resolution loss that had not been considered previously in terms of explicit MTF which can be grouped together with other system MTFs in performance simulation. We shall also describe artifacts which might appear as a result of line multiplications in imaging of time-varying scene. KEYWORDS: IR imaging, display MTF, Interpolation

Simple, recursive ray-trace equations are presented that exactly describe the propagation of a ray through a series of coupled, confocal, prolate spheroids. A vector representation is used for characterizing ray-propagation in different planes through the spheroid axis of symmetry.

Simple, exact ray trace equations are derived for a general ray propagating through tilted, coupled, confocal spheroids. In this off-axis optical system, there does not necessarily exist a (real) ray for which the paraxial approximation applies.