The paper presents the Clinton Administration's commitment to American industrial competitiveness through a strategic focus on research and development and to dual-use technologies in particular. Working in partnership with industry, the dual-use approach is essential for giving our armed forces the world's best, most technically advanced military equipment at affordable cost. The President has set a goal of shifting from a dominant role for military technologies in our Federal R&D investments to a roughly equal balance between military on the one hand and civilian and dual-use on the other. We have already made significant progress toward this goal.

Infrared thermal imaging instruments were originally developed by the military for passive night vision and surveillance and, subsequently, found applications in weapons aiming and target tracking. Commercial applications for this technology dictate different rules and place different emphasis on instrument configuration and performance characteristics. This paper begins with a performance comparison between military night vision systems and typical commercial imaging radiometers, including packaging, configuration and features. An overview classification of recent and current commercial applications follows, including a more detailed description of a few of the more unusual and innovative solutions. Finally, a glimpse of the near future is offered, based on the most recent developments in detector/technology, equipment packaging and imaging diagnostic software.

Wildfires in Europe are predominantly caused by people visiting natural areas. The hazards as a result of wildfires increase particularly in densely populated areas such as Europe, since more and more people visit natural areas. The circumstances for large wildfires to occur are predominantly located in countries surrounding the Mediterranean Sea. In these countries on average 1% of their natural area resources are affected by wildfires each year. Typically, a few percent of the total number of fires is responsible for more than 95% of the total area burned. The fire season is usually during the summer and early fall. The risks for wildfires in the northern European countries are generally much lower. Typically, 0.1 - 0.2% of the total natural area is burned each year. The main season for wildfires is early spring. The interest of the authorities, that are responsible for the natural areas, in the northern European countries for autonomous wildfire surveillance is high since the efficiency of the current wildfire surveillance, such as from airplanes and lookouts, is often unsatisfactory.

Preliminary analysis of the applicability of IR FPAs to monitor temperature changes during various laser therapies has been performed at the Beckman Laser Institute. Temperature measurements have been made implementing a high frame rate staring FPA. Extensive measurement of skin surface temperature during pulsed laser exposure of port wine stain was performed. Cursory experiments were performed on lung, teeth, and eye tissue during laser exposure.

Diffractive optical elements (DOEs) have been proposed for many applications. One of the principle limitations of these lenses is the abundant chromatic aberration that prohibits broadband use without design compensation. I propose a machine vision system that exploits this typically unwanted effect to provide information into a third dimension of current 2-D detection systems. This design implements a focal plane array with a diffractive focusing element. Output information can include the standard 2-D image, the distance to a component under test and if the object is in motion, the direction and velocity towards or away from the detector plane. This system can be designed to operate at visible or infrared (IR) wavelengths for a variety of inspection applications.

This paper describes a model for predicting the behavior of micro-machined thermal isolation structures that are employed in uncooled infrared detectors developed at the Defence Science and Technology Organisation (DSTO). The model forms the basis for calculations predicting the performance of semiconductor film microbolometers as infrared detectors. Performance predictions for a particular detector design are shown to be in good agreement with measured values. The effects of non-ohmic contacts and materials deposition techniques on noise, and hence on detector performance, are also discussed.

The time is ripe. Technological advances promise to permit civilian applications that are both cost effective and easy to use. The cold war is over and much of the enormous capability of the aerospace companies and engineers is available for peaceful exploitation. This discussion covers some of the technologies, applications, and policies that may and can facilitate the expansion of infrared technologies into the civilian economy.

A new multiscale wavelet transform algorithm for dim target detection in forward looking infrared (FLIR) imagery is described. The algorithm was evaluated using a database of FLIR image sequences representing different target, sensor, and background clutter scenarios. Detection performance was quantified in terms of the receiver operating characteristics (ROC) for all sequences. The detection performance of the new algorithm was compared to the Holmes double-gated filter method and was found to be comparable or significantly better depending on the image sequence.

In this paper the three-dimensional (3-D) restoration problem is addressed for multi- frame/multi-spectra space-based infrared image processing. The conventional restoration approaches are reviewed and new 3-D restoration methods are proposed. Issues related to multispectra restoration are also discussed. Experimental results are presented using sequential frames obtained from high altitude by an array of infrared sensors.

The Canadian Department of National Defence (DND) has established a requirement for a fleet of reconnaissance vehicles equipped with a modern surveillance system to be used in a wide variety of scenarios. This includes conventional operations within NATO, contingency operations in troubled areas as well as UN peacekeeping missions. As such, the Light Armored Vehicles Reconnaissance and Surveillance System will be the first 24 hour all- weather reconnaissance system integrated into a combat vehicle. This paper intends to describe how the operational requirements defined by DND were translated into sensor and system requirements. After a summary of the current configuration, it focuses on product pre-planned improvements and future needs.

The infrared target processor (IRTP) is a real-time processor capable of automatic and operator-aided detection, lock-on and tracking of multiple targets in infrared imagery. The IRTP has been developed under a feeder project to the advanced land fire control system (ALFCS) program, a project to develop accurate fire on the move capability for Canadian tanks. The IRTP is built around the datacube pipelined architecture using commercially available image processing and general purpose components. Presently, the IRTP is capable of real-time detection and tracking of up to five targets at video rates (30 frames/sec), and its operation is controlled via a single monitor that contains graphical control panels and a real- time video display. To augment the operation of the target detection/tracking and the gun fire control operations of the ALFCS a position orientation system (POS) can be used to provide accurate vertical reference measurements of the turret. The POS is built around an inertial measurement unit (IMU) and specialized real-time software implemented on a general purpose CPU card. Although the IRTP is currently being developed for ALFCS applications, the testbed architecture and algorithms are flexible and the system has been used to test and develop algorithms for general target processing applications.

Mirabel is a thermal sight developed for the Aerospatiale Eryx Weapon System. It is a `snap on' type sight which is attached to the weapon system's day sight. The system is lightweight (3 kg) with a recognition range of 1.2 km and boresight accuracy of 0.2 mR. Mirabel implements magnification one so it is not sensitive to interface accuracy between day and night sights. It provides for the use of the same eyepiece for both day and night usage, therefore allowing for human engineering optimization for both sights at once. The weapon system operates in both shoulder and tripod launch modes, and therefore imposes a very demanding human engineering consideration.

Properties of P(VDF-TrFE) copolymer have been studied. A characterization for thermal detection has been done on various copolymer samples with thicknesses ranging from 1 micrometers to 1 mm. a pyroelectric coefficient of 4 nC/cm² degree(s)C has been measured on samples polarized by the corona discharge method. Optical, dielectric, and thermal properties have also been measured. The possibility of polarizing the copolymer directly on an integrated circuit by corona discharge has been demonstrated and a monolithic detector array of 32 elements has been realized. Fabrication procedure, results of characterization and performance of the detectors are presented.

PbBiSrCaCuO films with a predominant Bi₂Sr₂Ca₂Cu₃O₁₀ superconducting phase were prepared on (100) MgO and LaAlO₃ substrates. The radiative properties of the film-substrate composites were investigated at room temperature and at temperatures near T_c in the wavelength band of 2 - 17 micrometers . For the film parameters used in this experiment, the near normal-incidence values of the reflectance were large while those of the transmittance were very small. No significant change in reflectance was observed as the film underwent a superconducting-normal transition in the vicinity of T_c. However, the reflectance decreased with decreasing film thickness and radiation wavelength. It could be verified that the substrate has a negligible effect on overall radiative properties when film thickness is sufficiently large. To determine the response mode of the film at infrared wavelengths, photoresponses to short laser pulses were measured. From the temperature- dependent behavior of transient structure and responsivity, the thermal origin of the photoresponse could be confirmed.

Some recent results on GaAs quantum well intersubband IR detectors for the long wavelength infrared are presented. These include studies of a systematic series of samples grown by molecular beam epitaxy (MBE) at three different facilities, and of multicolor detectors. In the first part, we compare performance of different detectors to address practical issues related to producibility using this new approach to infrared detection. We show that detectors made with specifications provided to several MBE facilities yield good results, which implies that these detectors are compatible with standard GaAs technology. We also show that the state-of-the- art GaAs-MBE layers are extremely uniform making this technology suitable for large focal plane arrays. In the second part, we apply the flexibility provided by the MBE growth technique in fabricating a voltage selectable multicolor detector.

This paper describes research, conducted at DREV, to make a proof-of-concept implementation of a real-time image stabilization and registration system for airborne long- range imaging sensors. The methods developed could be used for aligning image sequences in real-time or for decreasing bothersome image vibrations in airborne video. This requirement occurs for instance when images come from different sensors and have to be registered prior to applying data fusion methods. In other applications, image observation by human operators and/or computer vision systems is made difficult by the presence of mechanical vibration which mutually offsets sequential frames by either translational or rotational displacements. This may be the case when doing long-range observations with a helicopter-mounted camera where even small motion of the vehicle can result in appreciable image panning. The project had two aims: (1) the development and the optimization of algorithms and methods for image registration at the pixel and sub-pixel levels; and (2) the implementation of these solutions at video rate in hardware. This was accomplished using COTS (commercial-off-the-shelf) hardware on an experimental test bed. Pixel level alignment is done by using and adapting motion-estimation methods based on a block-matching algorithm (a discriminative selection of blocks for optimal motion estimation is done using an interest operator based on texture analysis). Sub-pixel alignment is based on inverse interpolation. In addition, a pyramidal scheme is used as a preprocessing step to improve the robustness and speed of the image- alignment calculations. Description of the hardware implementing the real-time execution of these algorithms is given. Finally, future areas of development and improvement are briefly discussed.

A total of 280 representative scenario profiles developed by Low and Hudak (1992) for the ARCTIC, MARPAC, CANLANT, and WESTLANT regions were extended in the vertical to 120 km. Their trace gas compositions, associated boundary layer conditions, as well as cloud and precipitation characteristics were examined for radiance and transmittance effects in the atmosphere by way of the LOWTRAN 7 model. Experiments with the LOWTRAN 7 cloud models were also carried out to investigate differences in average transmittances for the low, middle, and high clouds conditions affiliated with the profiles. Cloud-free line of sight frequencies for these regions were derived using the cloud data for each of the scenarios over a 10-year period.

A Canadian Naval threat/countermeasures simulator (NTCS) capable of modeling the engagement between a naval ship and an infrared (IR) guided anti-ship missile is presented. The NTCS program is built upon previously developed naval ship signature software entitled Ship Infrared Simulator (SHIPIR) which produces 3-D graphical imagery of a ship in its sea/sky background for a wide range of operational, atmospheric, observer, and spectral conditions. By adding models for an IR seeker head, missile flight dynamics and commonly deployed ship IR countermeasures, NTCS can effectively assess the IR susceptibility of naval platforms through calculation of target lock-on ranges and hit/miss distances. Current and future naval ships can be analyzed for IR suppression effectiveness in such areas as hot surface visibility, low emissivity paints, and engine exhaust signature suppression. The various deployable countermeasures (flares, smoke screens, washdown, and ship maneuvers) and missile/seeker heads modeled in NTCS permit the assessment of ship survivability and development of tactics and counter-measures necessary to provide adequate IR protection. A description of NTCS is provided with emphasis on the missile and countermeasure models and overall engagement simulation. Some sample simulations to date on the Canadian DDH-280 tribal class destroyer are presented.

An IR search and track (IRST) system is a system comprising a scanning IR sensor, a signal processor, and a track-while-scan track discrimination filter. The AN/SAR-8 IRST (a joint development of the U.S. and Canada) has undergone extensive test and evaluation in a naval environment. Trial data have been used to develop performance models for IRSTs. These models are described. The models ultimately assist in the synthesis of new IRST system designs.

Search and rescue (SAR) and maritime patrol missions pose a number of challenges for an imaging system. Systems must work in low light level, low visibility conditions to find and identify small targets for both search and rescue and law enforcement roles. Passive low light and thermal imaging systems are often unable to discriminate small targets against sea backgrounds due to low thermal contrast and non-cooperative targets. Active gated television (AGTV) as implemented in the ALBEDOS system, enhances the reconnaissance, surveillance, and SAR capabilities of maritime organizations by generating high resolution video imagery regardless of ambient light and conditions or target thermal properties. Active television uses a laser source to illuminate a scene being viewed by a low light television (LLTV) camera, can filter out unwanted light sources, and also limit the image depth of field. AGTV systems generate video for display or recording under conditions that are typically difficult for other sensors. AGTV systems have demonstrated their ability to provide long range detection of SAR targets, to allow the positive identification of people, and to read license plates and ship or aircraft markings covertly at long ranges. This paper summarizes the advantages of AGTV for reconnaissance and surveillance missions, briefly discusses the theories of operation, and compares AGTV performance to that of conventional sensors. A compact airborne AGTV configuration being developed for trials by the Canadian Forces is described.

This paper presents work performed at the National Research Council of Canada in infrared range imaging, i.e., from 1.5 micrometers to 1.8 micrometers . This region of the spectrum is chosen for eye-safety reasons. Basic concepts explaining the triangulation principle used in NRC prototypes are presented. The requirements for laser source and optics are described in some detail. Laser spot position detection is reviewed in the context of infrared range imaging with actual design examples and detailed calculations of signal-to-noise ratios. These calculations are useful in the early stage of a design. Experimental results show range images taken with the first prototype built at the Institute for Information Technology. A discussion on current developments concentrates on another prototype range camera intended for space applications. The current version of that prototype operates at 0.82 micrometers and can perform tasks in tracking mode at a refresh rate of 130 Hz (60 targets per second) or in imaging mode at a data rate of 18,000 registered 3-D and intensity points per second. The useful range is about 0.5 m to 10 m. For distances greater than 10 m, a time-of-flight unit along with a pulsed-laser source operating at 1.54 micrometers are included. The change of operating wavelength provides a system that is eye-safe and increases the signal to background light rejection for space applications.

A unique real-time hybrid optical-digital image processing system has been developed to perform analysis of low contrast visual and infrared images in radon-space. This system functions by implementing the forward radon transform (a mathematical tomographic transform of image data from two-dimensional image-space to one-dimensional radon-space) in a front-end optical processor, and a digital processing subsystem operating in radon-space instead of the more traditional image-space. The system works by optically converting the two-dimensional image-space data into a series of one-dimensional projections. All further processing is performed digitally in radon-space on the one-dimensional projections. Applications of interest such as the objective minimum resolvable temperature difference measurement of thermal imagers and automatic pattern recognition are discussed. Also, this paper discusses the topic of real-time object-moment analysis in radon-space which can be used for target identification under conditions of certain image distortions (size, rotation, translation, and contrast). Radon-space object-moments can be calculated using significantly less image data and fewer digital operations than those in image-space.

The SWIR full spectrum imager (SFSI) is an imaging spectrometer, covering the short-wave infrared (SWIR) from 1200 to 2400 nm, which has been developed for remote sensing from an airborne platform. The sensor has been designed to acquire the full spectrum at high spectral resolution (10 nm) and the full image swath at high spatial resolution (50 cm) simultaneously. The instrument utilizes a platinum silicide (PtSi) detector array, refractive optics, and a transmission grating. A VME bus computer communicates with the array controller, performs the data acquisition, and provides the operator interface. The camera and data acquisition subsystems have been completed and test flown. The fore-optics, spectrograph, and sensor housing have been fabricated. Integration of the camera, spectrograph, and auxiliary components is scheduled for July 1994 followed by laboratory testing and calibration. Our goal is to obtain pilot project data by the end of autumn 1994. Here we describe the optical design, the sensor system, early test flight image data, and expected sensor performance based on laboratory testing. The objectives and procedures for the spectral, geometric, and radiometric calibration of this sensor are also discussed.

A radiometer has been designed that operates at 1064 nanometers using a diffractive element arrangement to focus the energy onto a detector array. The aperture is made up of several elements consisting of both on and off-axis designed elements arranged to provide an overall FOV. The blur circle and the efficiency of the elements have been measured. The advantages of DOEs are weight saving, repetitiveness of design, the making of either off-axis or on-axis elements with the same ease, good efficiency of energy collection, and diffraction limited focusing.

In this paper, we present four diffractive objectives for a 240 X 320 pixel PtSi staring array. The focal lengths for these lenses range from 12.5 mm to 100 mm and they have a relative aperture of F/2. Various system constraints and their impact on the optical designs are addressed. In addition, measured and predicted performance data are compared.

Focal plane arrays allowed tremendous improvement in the robustness and compactness of thermal imagers reducing both mechanical and optical requirements. However, these will always be limited by the pixel size, the fill factor, and by the sampling theorem. As compared to older one-detector scanning systems, focal plane arrays can only reproduce half the frequencies scanning systems do for a given instantaneous field of view. To overcome this limitation, microscanning seems to be a winning approach. Microscanning can be seen as an oversampling process. A series of images representing the same scene are taken while displacing each time the image over the array by a fraction of the detector pitch. The oversampled image is built by interlacing all the pixels from all the images in both directions. It can be shown that microscanning can bring the resolution to the same level it is with standard scanning system. Furthermore, by characterizing the process, one can compensate for it and bring the resolution to the level of a microdisplacement. This article describes work that has been undertaken at the Defense Research Establishment Valcartier to evaluate the requirements for the microscanning process and to determine gains that can be obtained by using that technique in a surveillance application.

A hybrid system is proposed for automatic target recognition. The system developed consists of an optical module and an electronic/computer module. The electronic module uses a Datacube^TM image processing system to carry out real-time filtering, detection, and segmentation operations. The optical module is composed of two LCTVs (liquid crystal television), a He-Ne laser, a collimator, and a CCD camera. This module calculates specific energy spectra in the Fourier plane of a potential target identified by the electronic module. These values are fed in a neural network implemented on Sun SPARC 10^TM model 41 for the classification of the current target. Experimental results involving special classes of target and their distribution in the features space are presented.

The state-of-the-art in infrared optics can be looked at in terms of size, quality, material or complexity. Itek has recently completed the manufacture of an optical component that presented most of these challenging requirements to the designer, the manufacturing group and the test engineer. We believe that the combination of severe asphericity, test complexity and lightweight construction are representative of the most difficult optics producible today. The mirror is a convex asphere, with only bilateral symmetry, a departure from the nearest sphere of 178 micrometers , and a maximum slope departure of 4.0 micrometers per mm. Testing required a combination of null correctors and binary optics, with extremely tight alignment tolerances. Surfacing was accomplished with small tools and computer controlled optical surfacing.

The first-ever operational infrared focal plane that makes use of an architecture which mimics that of the human retina was designed, built, tested and integrated into a demonstration system at Amber Engineering. This paper presents an overview of the design concepts of this device, presents some operational results using the overall system, and discuses some options for system-level uses of future devices which will grow out of our initial experiences.

In the first part of this paper, a brief tutorial review of sensor fusion for target recognition applications is presented. In this context, relevant aspects of system architecture, sensor integration, and data fusion are discussed. Several unresolved issues in the practical implementation of sensor fusion are identified; significant among these are the rationale for selection of a sensor suite and a means for optimal allocation of requirements between sensors. In the second part, a novel methodology offering the potential for resolving these two issues is presented. An approach to defining sensor data information content is described. Coupling this approach with accepted rules of thumb describing target recognition capabilities results in a quantitative method for comparing the target recognition ability of diverse sensors (imagers, non-imagers, active, passive, electromagnetic, acoustic, etc.). Extension to describing the performance of multiple sensors is straightforward. Application of the technique to sensor selection and requirements allocation is discussed.

For over two decades, researchers have investigated a wide variety of technologies for use as a real-time infrared scene generator. During the past several years, the most promising technology to meet the myriad of applications appears to be the silicon micromachined resistive-array approach. Each thermal pixel is created by a micro-scale resistor. The present investigation reports the results achieved by using the standard commercial CMOS foundry process, rather than a costly custom CMOS process, to produce the chip and the subsequent post-foundry etching. Both chip-level and pixel-specific electronics are readily included on the chip since IC technology is employed. Principles used in device architecture formulation, chip design, and fabrication of large arrays of these thermal pixels are discussed. Experimental results of recent array designs are presented and illustrate that low-cost, high-performance flat- panel thermal infrared displays are now viable and practicable.

Gains in micro-optic technology may provide enhanced performance for IR sensing applications. The benefits in noise reduction and increase in signal-to-noise ratio on the detector arrays can off-set the increased cost of adding micro-lens structures to the detector assemblies. Additionally, new manufacturing techniques make it feasible to make micro-lens structures on the same substrate as the detector elements. One of the advantages of this technology growth is the shifting of alignment to the fabrication stage instead of the filter assembly stage. Important considerations include: fill factor, diffraction efficiency, optical and electronic crosstalk, optical power, and optical bandwidth.

Material selection and suitable low cost manufacturing processes for production of rugged man-portable lightweight optical test set components are described. Field requirements for an ultra lightweight optical test set comprised of an off-axis two mirror collimator plus source and detector optics mandated selection of extremely stable materials. The requirements include temporal, thermal and mechanical performance suitable for portable and transportable military applications. The environmental stability requirement includes operating over a temperature range of 130 degrees fahrenheit. Also military shock and vibration requirements for transportable equipment are imposed on the entire test set. The total weight budget is 5 pounds for the mirrors and the large supporting structure. The structure volume is only about 1% of the occupied space. A near-net fabrication process such as casting or HIP fabrication was required. A comparison of materials and manufacturing methods has resulted in the selection of a hypereutectic aluminum alloy containing 23% by weight silicon with stabilizing elements. This material can be cast and heat treated to produce uniform properties at low cost. The coefficient of thermal expansion (CTE) is much lower than other aluminum alloys and the modulus of elasticity is 50% higher. The alloy was machined with conventional tools and plated with nickel phosphorous of the same CTE to produce stable optics.

Focal plane array research is constrained by the fact that versatile, modular test stations are not available commercially. Typical imaging detector tests mandate fabrication of expensive unique image coupling devices simulating the actual systems under consideration. Stray light and field of view are often not well controlled during tests, leading to lowered signal-to-noise levels and ambiguity regarding performance of an actual system. The University of Alabama in Huntsville has developed an inexpensive process for producing image coupling devices suitable for testing and developing cooled focal plane array systems. Model image couplers for cryogenically cooled focal plane array detector research have been produced by this process. Electrodeposition procedures and precision single point diamond machining, combined with optical design procedures permit fabrication of unique optical systems. Aspheric designs are used permitting excellent control of distortion. Applications include laboratory and field studies of beam expanders, broadband and thermal image couplers suitable for non-contact measurement, IR spectroscopy and durable lightweight telescopes which image on focal plane arrays. Low cost deployable systems are also candidates for this technology.

Triglycine sulfate (TGS) is one of the most comprehensively studied ferroelectric materials for infrared detecting devices. The discovery of new materials with more attractive properties together with a growing requirement for uncooled thermal detectors, having better performance than the thermistor bolometers, has resulted in considerable attention to pyroelectric devices making them the most widely studied of infrared detectors. In recent years, the study of pure and doped TGS crystals has increased because of their promise in various devices. These devices have a number of important characteristics, such as low cost, low power requirement, and a wide operating range of temperature and frequency response. Pyroelectric infrared detectors find their use in military systems, astronomical telescopes, earth observation cameras, laser beam characterization, environmental analyses monitors, medical vidicons, and Fourier transform infrared (FT-IR) instrumentation. The paper describes the research done at Alabama A&M University on the growth of TGS crystals for infrared detectors. Also results of infrared detectors fabricated from TGS crystals grown in space on the First International Microgravity Laboratory (IML-1) and Spacelab-3 (SL-3) are described.

We developed a technique for growing Hg_0.7Cd_0.3Te on CdTe/sapphire. Using metal organic chemical vapor deposition (MOCVD), we grew CdTe on sapphire substrates. We use a combination of isothermal vapor phase epitaxy (ISOVPE) followed by liquid phase epitaxy (LPE) to grow Hg_0.7Cd_0.3Te on them. The ISOVPE converts the surface of the CdTe layer to Hg_0.7Cd_0.3Te, which decreases the lattice mismatch between CdTe and HgCdTe and decreases the surface defects on HgCdTe that originate from lattice mismatching. After growth, we annealed the wafer in a Hg atmosphere to control the carrier concentration. We used the performance of photovoltaic detectors to examine the wafer quality. A typical diode with a 5-micrometers cutoff wavelength had a responsivity (Re) value or 2.87 A/W. The diffusion current limits R₀A down to 120 K, and the generation- recombination (g-r) current limits R₀A at 77 K.

To meet the sensitivity and resolution requirements of high-performance long-wavelength infrared (LWIR) imaging systems, we developed hybrid HgCdTe 480 X 2 infrared focal plane arrays (IRFPAs) for the 8 - 10 micrometers band. We connected the hybrids using indium bumps and a sapphire wiring substrate to reduce the thermal expansion mismatch between the silicon readout circuits and the photodiode arrays. Using the mature liquid phase epitaxy (LPE) technology and a CdZnTe substrate, we fabricated LWIR photodiode arrays. Each photodiode array consists of 240 X 2-element n+/n/p diodes formed by boron implantation. The arrays have an average zero-bias resistance of 3.8 M(Omega) and a shunt resistance of more than 100 M(Omega) for a 10.5 micrometers cutoff wavelength. For the readout devices, we used n-channel charge coupled devices (CCDs) with charge capacities greater than 4 X 10⁷ electrons, and 4 signal outputs capable of 6 MHz data rates. The input stages of the CCD include skimming and partitioning functions. Operating at 80 K, the arrays had a mean laboratory D*_(lambda p) of 6.4 X 10¹⁰ cmHz^0.5/W with f/1.2 optics. The detectivity variation ((sigma) /m) was 14%, and the operable pixel yield exceeded 99%.

In this paper we present a staring IRFPA including an in-pixel dc current suppression scheme. This circuit allows a self-calibration mode (SCM) suppressing the dc current. A 64 X 64 IRFPA with 100 micrometers pitch and 4.2 micrometers cut-off wavelength MCT PV detectors has been achieved and tested at 200 K. At this temperature, the ratio between dark and useful photonic current is close to 100. After SCM, the residual integrated current is reduced to typically 1% of the detector initial current. This enables the integration time to be increased by two orders of magnitude, which drastically improves IRFPA sensitivity and resolution. At 200 K operating temperature, we obtained a peak wavelength responsivity (R_(lambda p)) of 10¹⁰ V/W and a noise equivalent differential temperature (NEDT) (temporally noise limited) of 80 mK (30 degree(s) FOV, 300 K background, 200 Hz frame rate). The SCM circuit avoids integration and readout of useless information and thus releases the signal-to-noise ratio (SNR) required for the readout integrated circuit (ROIC). This makes circuit design easier, for both multiplexers and output stages, and reduces the number of bits required for digital image processing.

Sofradir has developed a series of products by using the second generation IRFPA technology developed by the French research laboratory Lir with the support of the French MoD. This technology is based on the use of MCT photovoltaic diodes coupled to silicon readout circuit thanks to indium bump technique. The maturity is so good that producibility of such high performance, high complexity components was demonstrated several years ago. As a consequence, Sofradir produces now in quantities large IRFPAs sensitive in the 8 - 12 micrometers waveband. Beyond this success, Lir and Sofradir continue to update the technology and to build new products. Developments are on progress for long monolithic TDI arrays 3 to 5 and 8 to 12 micrometers , 480 X 4, 576 X 4 or staring arrays using CMOS or CCD readout circuits. All the work is performed within the frame of producibility and reliability.

Sofradir IRFPAs using MCT material are generally cooled down with classical coolers such as regulated or non regulated Joule Thomson coolers, rotary or linear closed cycle Stirling machines. By using these coolers, it is possible to satisfy a large range of application requirements. Nevertheless, the requirements for new military equipment increased during the last two years putting more pressure on cooling system requirements. On the other hand, the continuous increase of the size of available IRFPA also puts more pressure on cooling system requirements. Finally, the reduction of military budgets and the potential commercial market forces one to propose new configurations. Therefore, Sofradir has worked in the following different fields: (1) development of a very small flat detector dewar using a flat Joule Thomson cooler for severe environmental conditions such as seeker applications; (2) development of a detector dewar using the IDCA (integrated dewar/cooler) concept for high performance, and low cost application; and (3) development of detector dewar using thermoelectric coolers for low cost application. Sofradir presents the results of these activities.

A 128 X 128 element thermopile infrared image sensor has been developed. This device has a monolithically integrated structure to increase fill factor. The CCD for signal charge accumulation and signal charge read-out is fabricated on the silicon surface. Over the CCD, silicon dioxide diaphragms for thermal isolation are made by using micromachining technology. On each diaphragm, 32 pairs of p-type polysilicon and n-type polysilicon thermopile are formed. The noise equivalent temperature difference obtained by the device is 0.5 degree(s)C with an f/1 lens. Since the materials used are the same as those for silicon ICs, and since the whole fabrication process is carried out at the silicon IC plant, it can be said that a low cost uncooled infrared image sensor is realized by this technology.

A high-performance IR sensor has been developed at the Norwegian Defence Research Establishment (NDRE) as part of a test system for modern electro-optical fire control techniques. Diamond turned, aspheric optical elements and modern IRCCD detectors are used. Sensor output is available both as analog video and in digital form for connection to a digital image processor. Horizontal field of view is variable from 3.3 degree(s) to 60 degree(s) with a fixed spatial resolution of 0.2 mrad and a thermal resolution of better than 0.1 degree(s)K. The sensor is stabilized internally in roll and pitch.

For a wide variety of applications, imaging infrared seeker allows new generation weapon systems to reach high performance in terms of accuracy, standoff capability, target/false target discrimination, target recognition, etc. In the design of the seeker, the detector is a key component because the global performance is strongly related to the detector's and also because it has a large impact on the opto-mechanical concept. Thanks to the use of state-of- the-art CMT 2D IRFPA, new generation imaging infrared seeker (IIS) has very high performance (sensitivity, resolution, number of pixels) as well as a simple opto-mechanical architecture. Thomson-CSF Missile Electronics Division (DEM) has designed, manufactured, and tested breadboards of IIS with the support of French MoD. These IIS are based on 2D IRFPA provided by Sofradir under DEM requirements. The IRFPA selected for this development is a 128 X 128 CMT IRCCD sensitive in the 3-5 micrometers waveband integrated in an operational dewar cooled by a Joule-Thomson cooler. In this paper, first a technical description of the IR detector is given, then the electronic set developed is described, and finally the measured main figures are given.

This paper presents the InfraCAM^TM concept, key components, and test results for the standard production PtSi unit. In addition it shows test results using both InSb and HgCdTe arrays as they apply to InfraCAM.

New generation IRCCD focal plane arrays have pixel noise levels near BLIP performance. However, nonuniformities, nonlinearities, and 1/f noise can be limiting factors that affect the final performance of the system. The spatial noise introduced by these phenomena can be several tens of times greater than the temporal noise. Depending on the applications, the requirement to reduce the spatial noise below the temporal noise is always the same, but the dynamic range required by the system can lead to the selection of different means of nonuniformity correction. In the first section of this paper, we analyze the measurements of detectivities, nonlinearities, and 1/f noise for typical linear arrays of 288 X 4 elements used in the IRIS thermal imagers family. In the last section, results are presented about nonuniformity corrections, 1/f noise reduction by continuous updating of correction coefficients, two-point dynamic correction, nonlinear correction, and scene-based correction.

High resolution second generation infrared focal plane arrays (FPAs) are not yet fully mature. However, 480 X 4 infrared FPAs have been designed, developed, and demonstrated for the U.S. Army Line-of-Sight Antitank (LOSAT) Weapon System. The infrared FPAs were developed on schedule and in budget by Sofradir (Grenoble, France) by taking a low risk approach based upon the technology incorporated into their 288 X 4 production infrared FPAs. First pass processor success was achieved on the silicon readout integrated circuit, and concurrent engineering was used to optimize the infrared FPA for the particular requirements of the LOSAT program. The success of the 480 X 4 development has resulted in the availability of a high resolution infrared FPA that can now be produced for incorporation into forward looking infrared systems.

We suggest a method for spatially and temporally resolved IR-image detection by using a semiconductor-gas discharge (SGD-) structure. The operation of the device is based on the conversion of IR-radiation into the visible spectral range. Especially, we discuss the influence of the system parameters on the operation of the device and propose a technical realization for a real time IR-image conversion camera based on the converter cell in connection with a standard CCD-camera. In combination with a gateable CCD-camera or an electron-optical recording technique (e.g., streak- or framing camera), the converter structure could be used for high speed IR-imaging in the microsecond range.

Silicon VLSI technology can be used to build a high density readout circuit for the emerging high resolution IRFPA imaging system. The readout circuit has to be operating at liquid nitrogen (LN) temperature in order to both improve system performance and design simplicity as well as reduce complexity. Devices with various sizes are measured down to LN temperature. As silicon MOSFETs cooled down, various anomalies are observed. The observed anomalies in the measured results are described and explained whenever possible. The two most important temperature dependent parameters -- threshold voltage and carrier mobility are studied. Simple guidelines are provided to obtain suitable SPICE MOS level 3 dc parameters for circuit simulation.

The problems of dc restoration have bothered the designers of the thermal imager for many years. The distortion phenomena appear differently in different systems. The shape of distortion is dependent on the scan rate, the RC time constant or the width of the target. From the results of analysis, it shows that the MRT curve is quantitatively related to the time constant and scan rate; it also shows that dc drops are serious for low time constant or slow scan rate, and the 1/f noise is apparent in this case. From this analysis, the best conditions can be evaluated for different systems, and this method is helpful in designing a good system.

A new current readout structure for the infrared (IR) focal-plane-array (FPA), called the switch-current integration (SCI) structure is proposed and analyzed. It is found that the proposed SCI readout structure can achieve a good readout performance in a small pixel size. It is clearly shown through the analysis that the proposed SCI readout structure and the associated design technique can be applied to the high density and high performance readout circuit design. The high injection efficiency, good detector bias, high detectivity, and large storage capacity readout performances are achieved in a 50 X 50 micrometers ² pixel size.

Operation of the low-background IR extrinsic photoconductor detectors and detector arrays depends essentially on the process of spreading of the time-dependent photocurrent. The spreading occurs due to the low non-steady-state screening of electric charges in the detectors. It induces the next non-steady-state affects: the increase in the photoresponse time, the essential difference between the photocurrents in the circuits of the source contact injecting free carriers into the detector and of the contact which serves as a drain for the carriers, and the high cross talk in the detector array. The results of the experimental study of these effects and some other ones in 36- and 48-element linear arrays (Si:Ga) and their theoretical description are presented. The theory is in a satisfactory agreement with the experimental data. Some consequences of these results, which are important for the space-based-astronomy applications of extrinsic detectors, are discussed.

The capability to accurately predict the target acquisition performance of thermal imagers is extremely important for several reasons. The predictions can serve as a guide for weapons system development, and they can be used to decide if a particular design will satisfy established requirements. The predictions play an important role in war games and because of this they influence engagement tactics. They also aid in establishment of future weapons system requirements. Target acquisition performance prediction is also an extremely complex task. Several models of thermal imaging system performance have been developed over the years, many of them specialized to only a few military scenarios. The continued improvement in E-O technology combined with the need to predict acquisition performance for a wider range of scenarios have forced an evolution of the models. Systems analysts and model developers have been challenged in several areas related to model development. A brief review of the history of target acquisition model development is followed by an outline of some of the challenges we face at present. Some possible future directions are then discussed.

The effect of fixed-pattern noise on the detection of subtle targets embedded in Gaussian white noise was investigated, comparing human detection performance with that of an ideal linear observer characterized by the Hotelling Trace Criterion (HTC). Five signal-to-noise ratios (SNR) were tested for the two conditions: with and without fixed-pattern noise, where both sets of images also contained Gaussian white noise. In this study, the SNR was calculated using the Gaussian white noise only; the additional fixed pattern noise was not figured into the SNR. Human detection performance did not differ significantly in the two conditions. In both conditions performance increased significantly with increases in SNR. The HTC figure of merit predicted accurately the experimental results for the fixed-pattern noise.

For long range detection of low observable targets at sea infrared search and track systems are needed with high sensitivity. This goal can be achieved by integrating two-dimensional focal plane arrays into large optics. We have performed tests with such an integrated system in the 3 - 5 micrometers band, by using an InSb array with a 30 cm aperture. Besides this infrared band a scanning 10 micrometers detector is also placed in the optics for reference purposes. With the high frame rates (25 Hz) of present day FPAs we also have a powerful tool to reject background clutter (by temporal analysis). With this system we performed observations of atmospheric effects such as scintillation, under different atmospheric conditions. The gathered data were analyzed in a preliminary way, and both bands are compared.

To minimize the noise of a PtSi IR camera we examined all the sources of radiation reaching the detector. In particular, we suspected that the cold shield which is supposed to be (nearly) 100% black -- was not. In long wavelength FLIRs where the scene radiation is the main source of noise this was never an issue. In MWIR cameras this could be a source of added noise. The problem was to evaluate the noise contribution of the reflected radiation from the cold shield by sensor measurements. To this end we replaced the lens by an equivalent black (painted) tube which was temperature controlled (heated) and was outside the FOV of the cold shield. The mean charge in the center of the FPA was evaluated as a function of the temperature of the tube. The analysis of the test data showed that the cold shield reflectance is one percent, a low value. As utilized in our IR camera, this commercially available cold shield was determined to be effective in keeping the noise low.