This paper provides an overview of calibration requirements and calibration approaches for new U.S. Navy infrared support equipment. The new infrared systems that are being developed in the Navy will require more complex test stimuli than those currently in use. The requirements will include multispectral capabilities, multisensor boresight alignment (infrared sensor and laser receiver, microwave and infrared), and dynamic thermal resolution targets. Imaging systems will require testing of the image processing functions of the system. New and improved calibration equipment and methodology will be needed to meet these more complex test requirements. However, changing test methodology should not necessarily impact current calibration philosophy. Support for certain parameters at the National Bureau of Standards (NBS) will also require development or improvement.
Testing issues arising from producibility concerns have prompted the US Army Strategic Defense Command to rethink its traditional dependance on complete capability seeker/sensor test chambers. The entire testing spectrum is being reevaluated. Component or subsystem tests joined together by computer simulation to produce end item performance predictions can be used to increase throughput to expected production rates. This paper discusses these and other testing issues, their applicability to phased development of a seeker/sensor system, and concentrates on ideas for solutions to the production rate problems.
The primary focus of this paper is long wavelength infrared sensor calibration. The success or failure of a calibration program is determined long before calibration time by the quality of the prior sensor design and test program. Failure to adhere to any one of several important test principles can lead to difficulties in calibration and flight. If test and calibration are not given high priority during sensor design, flight performance may be compromised because calibration and functional testing will be difficult. A test and calibration facility is virtually required to accomplish the many different tests essential for sensor acceptance and calibration. A number of critical issues remain to be resolved, and a number of new requirements must be accommodated in future test and calibration facility design.
The National Bureau of Standards (NBS) has accepted the responsibility for constructing a new facility for the calibration of radiometric sources used in low background environments. Along with the facility's development, a program has been started to study the long term needs of the calibration community and to develop appropriate new test and measurement procedures. The facility consists of a large (60 cm dia by 152 cm long) vacuum chamber whose inner volume is maintained at temperatures of less than 20 K. The radiometer is separated from the source by an isothermal wall whose temperature is actively controlled. The cryogenic cooling of the vacuum interior is accomplished by a closed cycle helium refrigerator system which also contributes to the vacuum maintenance. The source volume can accept a cubic source of up to 30 cm on a side provided the aperture is suitably located. A research and development project is underway to develop methodology to characterize optical attenuators over many orders of magnitude and over a broad wavelength range.
The Characterization of Advanced Low-Background Mosaics (CALM) test chamber is a low-background, high-vacuum cryogenically cooled test facility used to test and evaluate long wave infrared mosaics focal planes. CALM is capable of generating broad band and narrow band sources, spot scan and flooded testing, spectral characterizations, and provides for quick turn around of FPA's. This paper describes the CALM chamber and associated support equipment and discusses representative test data from mosaic detector arrays.
This paper describes the implementation of a Seeker Evaluation and Test Simulation (SETS) Facility at Eglin Air Force Base. This facility will be used to evaluate imaging infrared (IIR) guided weapon systems by performing various types of laboratory tests. One such test is termed Hardware-in-the-Loop (HIL) simulation (Figure 1) in which the actual flight of a weapon system is simulated as closely as possible in the laboratory. As shown in the figure, there are four major elements in the HIL test environment; the weapon/sensor combination, an aerodynamic simulator, an imagery controller, and an infrared imagery system. The paper concentrates on the approaches and methodologies used in the imagery controller and infrared imaging system elements for generating scene information. For procurement purposes, these two elements have been combined into an Infrared Digital Injection System (IRDIS) which provides scene storage, processing, and output interface to drive a radiometric display device or to directly inject digital video into the weapon system (bypassing the sensor). The paper describes in detail how standard and custom image processing functions have been combined with off-the-shelf mass storage and computing devices to produce a system which provides high sample rates (greater than 90 Hz), a large terrain database, high weapon rates of change, and multiple independent targets. A photo based approach has been used to maximize terrain and target fidelity, thus providing a rich and complex scene for weapon/tracker evaluation.
New infrared calibration test facilities are required to provide for the calibration of sensors that exhibit resolution in the microradian range and that utilize a large number of detectors in linear and area arrays. Simple collimators with low divergent beams are typically physically large, costly to cool, and provide only a partial calibration. Other capabilities such as multiple point sources, bar patterns and an extended-area source are needed. A compact portable multifunction calibrator is designed for future sensor systems that enables a linearity calibration for all detectors simultaneously using a near small area source ( Jones source), a high resolution mapping of the focal plane with 10 prad setability and with a blur of less than 100 prad, system spectral response calibration (radiometer) using a Michelson interferometer source, relative spectral response (spectrometer) using high-temperature external commercial blackbody simulators, and an absolute calibration using an internal low-temperature extended-area source. In addition, a scatter plate is available to provide a diffuse full-field and full-aperture but attenuated high-temperature radiation source, bar pattern reticles to provide direct evaluation of modulation transfer function and and bandpass filters to provide system parameter evaluation at selected wavelengths. The portable system is made compact through the use of a folded Gregorian collimator design and includes an extended-area source, scatter plate, and Jones source which are stowed in the system and can be switched into the beam.
A rugged IR scene simulator (IRSS) is described which combines two, independently moving, dynamic IR targets and a third background source by means of a single beamsplitter. Each target may be separately controlled to produce pitch, yaw, and range closure motions under computer control. The entire system is designed for mounting on the inner axis of a high speed Target Motion Simulator gimbal system surrounding a three axis Missile Motion Simulator. Interchangeable optics for 1.5-5μm or 8-14μm may be employed or broadband 2.25-141μm optics can be used with some minor loss of resolution. Easy access to the two focal planes permits substitution of different servo controlled shaped targets or the use of dynamic pixel arrays interchangeably. Cooling of the internal optics permits operation with low background radiance from the system itself. Both the IRSS and the various target modules presently existing and planned are discussed.
Evaluation of proposed new U. S. Navy calibration methodologies and techniques will be performed using an imaging radiometer. Mr. Westfall and Mr. Larason develop an analytical model for initial analysis and present results of the first-order performance model developed during conceptual design. Scanner selection and the blackbody references are addressed, and the performance prediction for the detailed design is included. The paper then turns to noise versus encoding interval, discusses the system block diagram, and summarizes anticipated calibration and testing requirements.
In this paper we consider an axially symmetric blackbody (BB) measurement system with a circular aperture and a circular detector. The BB can be of a right circular conical shape, a right cylindrical shape, or a combination of these two shapes. Assuming that the BB is ideal, we calculate the power received by the detector. Using the axial symmetry we reduce the results to products of a single integral, involving Planck's distribution and the spec-tral response of the detector, and a double integral with variable limits of integration. The latter integral, called "system geometry factor (SGF)", depends on the geometry of the BB, the sizes of the aperture and the detector, and the distance of the detector from the aperture plane. When the aperture and the detector are sufficiently small and the distance between them is large enough, the SGF can be calculated as if the BB cavity were an extended source over the aperture or a point source situated at the center of the aperture. The SGF can be used in many situations. As illustrations, we show how the exact calculation and the approximations can be used for designing a test, and for checking the effectiveness of a BB or the whole measurement system. We also discuss the uncertainty relationship between the BB temperature and its radiant power measurements and how it is used in different practical cases.
This paper describes an improved Extended Area Blackbody Source for testing thermal imagers on the production line and in the field. Emphasis is given to the Dynamic MRT capability, and Scene Targets. The paper details work performed on this improvement program. Initial work was described in our previous paper "Microprocessor based radiation sources", Ron Fourier, et al, published by SPIE Volume 520, 1984. I will describe the tests performed on this source, including Calibration, Environmental temperature stability, Radiometric calibration and uniformity tests.
A multiple-celled extended blackbody source with two temperatures which has an operating temperature range of 40°C to 150°C, utilizing water/copper gravity heat pipe for uniform and stable temperature control, is reported in this paper. And its effective working area is 420x234 mr11. AGA-780 thermovision and THERMO TRACER 6T61 are used to measure the temperature distribution over the working area of the extended source and the results indicate that the maximum value of the temperature difference within ninety eight percent of the working area is less than 1.5°C. A method of calculating the integrated emissivity of a multiple-celled extended source defined by us has been used to compute the integrated emissivity of the extended source: The results show that the integrated emissivity of the extended source to a concentric detector is 0.993 when the apparent emissivity of the black paint is 0.98.
An extended blackbody source with cavity effect realized bx gravity heat pipe is reported in this paper, which has an operating temperature range of 40°C to 150°C. The diameter of its working area is 150 mm. THERMO TRACER 6T61 is used to measure the temperature distribution over the working area. The results show that the maximum temperature difference within ninety eight percent of the working area is less than 1.4°C. The integrated emissivity of the extended blackbody source to a concentric circular detector is 0.996. This extended source has advantages of relatively good temperature uniformity, simple construction, and low cost.
The calculations are presented of integrated cavity emissivities of diffuse cones, cylinders and cylindro-cones all of which have annular lids. The semi-visible angle factors of infinitesimal ring to coaxial disc have been developed into an analytical expression so that the calculations can be run exactly and simply.
This paper will consider the critical scene generator device characteristics and specifications, and then review the most prominent technologies; Bly Cell, Electrically Heated Pixels, Liquid Crystal light valves, Vanadium dioxide modulators and mirror matrix modulators. Each method will be critically examined for its ability to meet the various requirements.
The Air Force Armament Laboratory (AFATL) at Eglin Air Force Base, Florida is currently examining the technology for the design and development of an infrared (IR) target projection system. This system will become part of an infrared target simulator (IRTS) for supporting the hardware-in-the-loop testing and analysis of infrared imaging (IIR) seekers. Although other projection technologies exist, we believe the thermal emitter array technology to be one of the most promising. This paper will discuss some of the reasons for such a belief. The Bly Cell, the liquid crystal light valve, and the thermochromic light valve are some of the other technologies the Laboratory has examined. While each technology is characterized by its strengths and weaknesses, only the liquid crystal light valve and the thermal emitter array will be covered at length in this paper. Inherent characteristics such as speed-of-response, image resolution, and temperature dynamic range determine the ability of each of these technologies to accurately simulate IR air-to-surface and IR air-to-air targets. This paper will discuss the advantages and disadvantages of the liquid crystal light valve and the thermal emitter array.
This paper addresses the issues involved with the infrared scene generation in a cryogenic test chamber for testing IR surveillance sensors and interceptor seekers. To resolve the problems associated with the simulation of IR targets using thermal blackbodies, a simulated blackbody for IR scene generation is proposed, along with a discussion of its validity and advantages offered by this model.
This paper describes the concept and current status in the development of an infrared simulator for testing electro-optical (EO) sensor systems against smoke/obscurants. The objective of the research is to implement realistic indoor smoke/obscurant testing of a variety of military electro-optical sensor systems under highly repeatable, computer-controllable conditions. The concept is to replay appropriate, previously recorded imagery from free-air smoke/obscurant field tests, utilizing an infrared scene generator. A reduced set of requirements was developed by limiting the application to a HgCdTe-based sensor system, the tank thermal sight, which represents a class of FLIRs that might be tested using the simulator. The simulator's requirements in terms of the role of scattering, target spectral characteristics, dynamic range, time constant, and spatial resolution are discussed. A group of technical approaches to infrared scene generation and their relative attributes are briefly reviewed, and the technical approach selected for continued development is described. The status of the research and performance of the current proof-of-concept image generator are summarized.
The purpose of the Dynamic Infrared Background/Target Simulator (DIBS) is to project dynamic infrared scenes to a test sensor; as an example, a missile seeker that is sensi-tive to infrared energy. The projected scene will include target(s) and background. This system was designed to present flicker-free infrared scenes in the 8pm to 12μm wavelength region. The major subassemblies of the DIBS are the Laser Write System (LWS), Vanadium Dioxide (V02) modulator assembly, Scene Data Buffer (SDB), and the Optical Image Translator (OIT). The process of writing a scene begins when the SDB is supplied with scene data from an imaging processing system. After processing, the SOB sends the data to the LWS which writes the scene onto a pair of VO2 modulators. This scene is then projected through the OIT giving a 6° x 6° picture to the sensor being investigated. This paper describes the overall concept and design of the infrared scene projector followed by some detail of the LYS and VO2 modulator. Also presented are brief descriptions of the SDB and OIT.
A unique projector has been developed that projects a dynamic scene from successive images on film in both the visible and infra-red portions of the spectrum. The theory, construction, and performance of the projector, called the SCANAGON, is presented.1 It is truly flickerless, having a negligible illumination variation from scene to scene. Each successive image is "wipe-dissolved" into the previous image in a smooth transition by means of an unusual mirrored scanner. Used with the BLY cell, it projects a dynamic scene from motion picture type film into the visible wavelength side at up to 4500 foot-Lamberts, which will saturate the cell. A black-body source may also be used in conjunction with a film which modulates infrared radiation. The images can be programmed to flow at a rate from zero to several hundred frames per second, forward or reverse.
Military infrared systems are growing increasingly sophisticated. However, the methods required to test them are lacking. In order to effectively evaluate these systems, new techniques are needed for simulating real world infrared (IR) scenery. An investigation of a new approach toward the development of a dynamic infrared image generator is described. The primary goal was to demonstrate flickerless infrared scene presentation. The system is based on the use of a flickerless video projector and a visible-to-infrared transducer (VIRT).
In this paper we describe the design and performance of EHP arrays constructed at British Aerospace (BAe) for the purpose of dynamic infrared scene generation. True black body unpolarised infrared radiation is achieved by the individual control of resistor elements from an electronic backplane. Applications designed for to date have demanded a temperature range of 20 or 30°C, although the material limitations permit designs upto 400°C. Designed and achieved time constants in the millisecond range have been obtained, allowing demonstration of real time picture generation in hardware in the loop simulation systems. Videotapes of real time image generation at 100 x 100 will be shown.
An Infrared Dynamic Scene Generator (IRDSG) was developed at Ball Aerospace Systems Division (BASD) in order to test an integrated sensor/processor. Using halftone images silkscreened on the surface of a mirror, the DSG presents to the sensor a replica of the inband radiances of a selected scene, taken from a digital database. The mirror with the scene printed on it is mechanically moved across the sensor's field of view. The energy from a constellation of simulated point targets is added to the image by a tuned beam-splitter. Their position and motion are independently controlled. Target multiplication increases the number of simultaneous tests per run while allowing the location of extremely faint targets to be known a priori. The line of sight of the sensor is also controlled for the simulation of scanning routines and platform jitter. The system is simple to set up and operate, has a wide dynamic range and is inherently flicker free. We discuss the implementation of each of the subsystems including two processes for transferring the image from the digital database to the reflectance mapped mirror.
Current interest in space based interceptors using IR sensors to home on and intercept ICBM's during early phases of flight make earth background scene generation in the opaque, infrared bands between 3.0 - 15.0 (μm) an important issue. This paper describes an approach for generating scenes in these opaque regions for arbitrary geometry and spatial resolution from measured data of fixed geometry and limited spatial resolution. Power spectral densities (PSD's) in the appropriate regions are constructed from various measured sources and used to synthesize images for a specific geometry by fractal continuation. This approach is particularly useful since it avoids the severe computational complexity and time penalty involved with a first principles approach. This is especially important for generating the large number of scenes required for ground testing high frame rate KEW sensors. An evaluation of the sensitivity of a typical acquisition algorithm to the accuracy of the background PSD's used to generate the synthetic scenes is also presented.
A generic sensor simulation has been developed which emulates the imagery produced by closed circuit television sensors. Applications to date include a monochromatic vidicon and a color CCD camera. The core software program was extracted from the MARSAM II model1 embedded within the Avionics Laboratory Sensor Performance Model (ALSPM)2. The program is written in FORTRAN and runs on a VAX which hosts a Gould Image Array Processor. The latter is programmed at the driver level using LEVEL 0 commands to enhance execution speed. The entire program is very user friendly, with menu prompting and default parameters provided. The operator/analyst is provided the option at each of the processing steps of assigning a specific parameterization corresponding to the electro-optical attributes of the sensor. Such parameters can be entered on a trial basis for temporary evaluation, and then discarded if found to be unsatisfactory. Such changes are made on-line in an operator interactive mode, thereby making the simulator a powerful design tool.
Currently there is a large ongoing effort to develop strategic defense weapons to defend against a ballistic missile attack. Before these systems are deployed in the field, there must be great confidence in their ability to perform their assigned task. Strapped down closed loop simulations provide the most economical method of achieving this confidence. Creating the imagery for these simulations will be a major effort. This paper reviews the requirements of Computer Image Generation (CIG) systems used to produce imagery for these simulations.
A capability for generating sky images containing both clouds and aircraft has been developed through concatenating a variety of models and matting them to an Image Array Processor. A family of apparent images are generated, each at a specific wavelength, whose gray scale values are in terms of absolute radiometric units. These images can then be combined as weighted sums to represent the appropriate spectral distributions for a specific sensor of interest. For example, this simulator has been used to generate sky scenes as inputs to a color CCD camera emulation.
The paper describes an all-electronic multi-channel waveform generator capable of simulating the multiplexed analog signals that leave the focal plane of a scanning mosaic infrared sensor. Target scenarios, background clutter, and various noise components are computer loaded in near real-time and the all-electronic simulator accurately synthesizes the many detector responses, and then outputs the sampled detector information in a format representative of the sensor being simulated. To facilitate rapid computer loading, each scenario object is specified by its amplitude and two dimensional position in the far-field and logic circuitry within the simulator deduces the multiplicity of detector waveshapes that result as the optical blur spot scans across the defined focal plane detector layout. Provision is made to handle dense target clusters and the initial simulator will be capable of simulating 5120 detectors sampled at typically 5000 samples per second. The analog detector samples are output over eighty separate D-to-A convertor channels, each carrying time-multiplexed signals from one or more columns of detectors.
Figures of merit for the unbiased comparison of in-band radiant intensities or irradiances to model predictions or observations of the identical fluxes by a second sensor have been developed. For Gaussian errors the figures of merit obey x2 statistics for the appropriate degrees of freedom. Noise sources rigorously treated are: 1) time-band dependent signal-to-noise (S/N), 2) in-band precision (σ), 3) uncorrelated bias or band-to-band error (ε), and 4) correlated bias (A). The figures of merit are simple to apply because all required matrix inversions have been performed analytically. The figures of merit are based on a group theoretical-like decomposition of the Mahalanobis distance where the features are radiant intensities and sensoronly uncertainty contributes to the covariance matrix. The principal application of these formulae is expected to be model validation, although past related applications have been various.
The generation of real-time dynamic infrared (IR) scenes is a fundamental problem when evaluating current and planned IR tracking systems. This problem has been the subject of a number of studies and a tri-service working group. The Dynamic Infrared Missile Evaluator (DIME) at the Air Force Wright Aeronautical Laboratories (AFWAL) must be able to simulate real-time dynamic IR scenes/targets, if it is to stay current with advanced IR missiles and evaluate countermeasure techniques against them. This paper discusses an approach which will give the DIME this dynamic IR scene simulation capability. The approach selected is the result of a study of many alternatives by Quest Research Corporation under a task entitled, "Dynamic Infrared Missile Evaluator Upgrade." The advantage of the selected approach is the ability to generate an IR image which is flicker free and does not require special high speed, real time digital image generation. The key component of the system is a Hughes liquid crystal light valve (LCLV) which converts a visible image into an IR image. A visible transparency is made of the desired IR scene and projected onto the LCLV. Scene motion is simulated with an x-y translation of the transparency and a 100/1 computer controlled zoom lens. The transparency will be generated in house using an image processing system. This system will allow the visible transparency contrast to be adjusted to precompensate for the optical transmission characteristics of the visible and IR portions of the simulator so that the true inband contrast is presented to the seeker under test. Limitations of this approach include a maximum seeker-to-target range variation of 100/1 and the inability to vary the target/scene aspect during a missile flyout.
A dual color calibrated imaging radiometer is being developed by Magnavox under government sponsorship to provide background and signature data and ground truth for the Automatic Target Recognizer (ATR) community. The system features high spatial and thermal resolution to be consistent with second generation imaging systems using currently available technology. Special design features are incorporated to yield accurate apparent temperature readings in both the 3-5 and 8-12 micron regions with pixel to pixel registration. The system consists of a fully militarized sensor head and remote processing electronics containing a mixture of customized and commercial processing equipment. The electronics may be up to 150 feet from the sensor head. The system output consists of two simultaneous video signals, one for each color. The videos are available in both an RS-170 format and an ATRWG digital format. The system produces 12 bit video. The video of each channel is analyzed by a dedicated microcomputer to provide real time data reduction to the operator.