The three emissive technologies--thin film resistor, silicon bridge resistor and suspended membrane resistor--currently being investigated for dynamic infrared projection applications are thermally compared and contrasted. It is shown that the technologies may be compared quantitatively in terms of both the strong power/speed trade-off that exists and the maximum effective blackbody temperature capability. The suspended membrane technology is shown to be superior in thermal terms, followed by the thin films technology which is found to be practical for high speed small array applications, or for large array applications provided that only small levels of effective blackbody temperature are required. In comparison, it is found that the capability of the silicon bridge technology is constrained in thermal terms both by the comparatively high thermal conductivity of silicon and by the small values of fill factor inherent to the technology.

We describe an experimental program of design, fabrication and preliminary performance of an infrared scene generator system designed as a 256 X 256 element electrically heated pixel array. Each pixel consists of a suspended resistor element which has a high area fill factor, being designed as a 3D structure of an electronic drive backplane. Initial performance shows that a practical device can be designed with around 300 degree(s)C temperature range.

A methodology for the simulation of infrared scenes is presented. The approach is based on combining synthetic and real data. The thermal signature of each object is first estimated using a thermal model. Environmental parameters which contribute to this signature are also considered. Then, specific techniques are used to extract textures from real images and to use these textures in the generation process. Experimental results showing purely synthetic images and synthetic textures images are presented.

Honeywell Inc. and Mission Research Corporation (MRC) are jointly developing micro resistive heater array displays for projecting dynamic background scenes and targets in the short-wavelength infrared (SWIR) to long-wavelength infrared (LWIR) wavebands. There are two joint government contracts supporting this work: the Nuclear Optical Dynamic Display (NODDS) program under DNA contract DNA001-92-C-0041, which is developing a 512 X 512 array of 50- by 50-micrometers display pixels, and the Cryovacuum Resistor Infrared Scene Projector (CRISP) program for the USAF Wright Laboratory at Eglin AFB, which is developing a 512 X 512 array of 87.5- by 87.5-micrometers pixels. The requirements on the two programs are somewhat different due to their different missions. While the NODDS program is developing an array that can be used to create dynamic nuclear clutter scenes, the CRISP arrays are being designed for simulating multiple independently moving targets; and while the frame rate on the NODDS arrays requires an array capable of 1-kHz frame rates, the CRISP arrays will be operated at 30 Hz.

The Strategic Defense Initiative (SDI) must simulate the detection, acquisition, discrimination and tracking of anticipated targets and predict the effect of natural and man-made background phenomena on optical sensor systems designed to perform these tasks. NRL is developing such a capability using a computerized methodology to provide modeled data in the form of digital realizations of complex, dynamic scenes. The Strategic Scene Generation Model (SSGM) is designed to integrate state-of-science knowledge, data bases and computerized phenomenology models to simulate strategic engagement scenarios and to support the design, development and test of advanced surveillance systems. Multi-phenomenology scenes are produced from validated codes--thereby serving as a traceable standard against which different SDI concepts and designs can be tested. This paper describes the SSGM design architecture, the software modules and databases which are used to create scene elements, the synthesis of deterministic and/or stochastic structured scene elements into composite scenes, the software system to manage the various databases and digital image libraries, and verification and validation by comparison with empirical data. The focus will be on the functionality of the SSGM Phase II Baseline MOdel (SSGMB) whose implementation is complete Recent enhancements for Theater Missile Defense will also be presented as will the development plan for the SSGM Phase III Operational Model (SSGMO) whose development has just begun.

A Scene Generation Test Capability (SGTC) is under development at Arnold Engineering Development Center (AEDC) which uses Direct Write Scene Generation (DWSG) as a tool to project realistic mission scenarios into sensors operating in a simulated space system environment. This capability can reduce the risk associated with developing advanced sensor systems. The second phase of this program, a Focal Plane Array Test Chamber (FPATC), is currently underway which expands the capabilities of the phase one Transportable Direct Write Scene Generator (TDWSG) reported previously. Projection wavelengths for the system include 0.514, 1.06, 5.4 and 10.6 micrometers . Multiple chamber configurations will be used to accommodate different types of test articles. The FPATC is also transportable. User testing has begun in the TDWSG. This paper will present an overview of the current SGTC program.

A novel concept for dynamic IR scene projection using IR diode lasers has been developed. This technology offers significant cost and performance advantages over other currently available projector technologies. Performance advantages include high dynamic range, multiple wavebands, and high frame rates. A projector system which utilizes a 16-element linear array has been developed and integrated into the millimeter wave/infrared (MMW/IR) hardware-in-the-loop (HWIL) facility at the US Army Missile Command's (USAMICOM's) Research, Development, and Engineering Center (RDEC). This projector has demonstrated dynamic range in excess of 10⁵, apparent temperatures greater than 2500 degree(s)C, and nanosecond response times. Performance characteristics for this projector system are presented in the paper. Designs for projectors to test other IR sensor configurations, including FPAs, have been developed and are presented as well. The FPA design consists of a linear array of diode lasers scanned by a polygon mirror. This low-cost projector offers high resolution, high contrast 2-D scenes at up to 10 KHz frame rates. Simulation of active IR countermeasures is another promising application of diode laser projector systems. The diode laser is capable of simulating flares or virtually any IR jammer waveform.

Under the Seeker Exo/Endo Designation Demonstration (SEEDD) Program, a multi-focal plane array sensor was interfaced to the Kinetic Energy Weapons Hardware-in-the-loop Simulator (KHILS) Facility to evaluate real-time tracking and discrimination algorithms using simulated infrared signature data of reentering objects. This simulation, as is typical with most HWIL simulations, consisted of two main subsystems: (1) the KHILS scene generator/projector system, and (2) the optical unit under test (UUT) including its associated electronics, processors, and software. The KHILS system presents two-color overlaid 96 X 96 laser addressed object scenes. The SEEDD sensor consists of a three-color catadioptric infrared system incorporating scanning and staring systems. Accurate simulation of flight events requires the thorough understanding of these systems, both individually and collectively. Using the KHILS and SEEDD systems as an example, a preliminary discussion of the issues associated with the interfacing of these types of systems is given.

Hardware-in-the-loop (HWIL) testing can be used as an efficient and effective means for analyzing the performance of guided missile systems. Due to the limits of current technologies, components of the simulation are limited in their capability to simulate real-world conditions for certain test articles. One component which is critical in an HWIL system for strategic guided missiles is the scene projection or delivery device. To stimulate imaging JR sensors, this scene projector (SP) typically consists of a pixelized in-band source which can be modulated both spatially and temporally to simulate the radiane scene which would be observed during an actual engagement. The SP is driven by a scene generator which provides scene radiance information to the SP under control of a simulation computer, which determines the field-of-view (FOV) composition based on a simulated engagement. In using such a system, a primary concern is that the SP is able to create a scene which produces the proper response in the observing sensor. Another effect which bears examination is the SFs projection method, such as scanning an in-band source to cover the projection FOV. The detailed interaction between the modulated source and the timing of the sensor's detection, integration, and readout processes may cause unrealistic or unexpected sensor behavior. In order to assess the compatibility of a specific sensor viewing a specific SP, a detailed simulation has been developed by Nichols Research Corporation under the direction of the Guided Interceptor Technology Branch (WL/MNSI) of the USAF Wright Laboratory Armament Directorate. This simulation was designed primarily to address issues related to scene projector usage in the Kinetic Kill Vehicle Hardware in the Loop Simulator (KHILS) facility at Eglin AFB, Florida. The simulation allows the user to define: the spatial response of the sensor; the spatial properties of the SP (i.e. the radiance distribution arising from a commanded impulse); the illumination timing of the SP, such as scan format, persistence, etc.; and the integration and readout timing of the sensor. Given sampled values of these response functions, and sampled values of the desired radiance scene, the SP simulation computes the detector outputs in the form of a sensed image. This output image can help to assess the suitability of using the modeled SP for testing the modeled sensor by illustrating potential mismatches. It also provides a means to predict the performance to be expected from this module of the HWIL simulation for a particular test scenario. This paper derives equations which express the sensor output as a function of the input scene, the spatial and temporal response functions of the sensor and the SP, and the spectral response functions of the sensor and SP. Assumptions which affect the implementation and the generality of application are stated and discussed. Results and conclusions are presented for a specific application which illustrate the utility of the simulation

This report proposes a generic methodology for generating infrared signatures of airborne targets. The various issues, assumptions and simplifications utilized in signature studies are outlines to insure baseline consistency among future models and evaluation tools. More specifically, the target is characterized on a component level, and the at-aperture signature is generated by the correct inclusion of atmospheric transmission. While the technique and general concepts may apply to all airborne targets, this study places emphasis on cruise missiles and related targets due to their low contrast. For these targets, the background signature becomes more important as both the emitted target radiance and the reflected background radiance contribute to the overall signature. Example target signatures generated using the proposed methodology will be presented following the discussion of signature modeling.

In Direct Write Scene Generation (DWSG) a critical lens design parameter is the optical crosstalk, or the undesired energy incident on adjacent pixels (EOAD). Crosstalk must be minimized to obtain the proper scene projection fidelity. The laser scan lens systems used in this scene generation technique are typical f(theta) and can be optimized by adjusting the diameter of the input gaussian beam in relation to the entrance pupil. The optimum gaussian beam diameter is dependent on the resolution requirements placed on a system and the test article metrics. Optical cross talk calculations are used to gauge the performance of systems with square and circular entrance pupils. This paper will address the calculated relationship between crosstalk and the truncation of the Gaussian input beam.

Sampling is present in all electronic imaging systems. For scanning systems, the scene is sampled in the cross scan direction by the discrete location of the detectors and by the A/D converter in the scan direction. For staring arrays, the discrete location of the detectors samples the scene in both directions. Sampling creates both phasing effects and aliasing. Since the aliasing occurs at the detector, it cannot be avoided. After a signal has been aliased, it cannot be reconstructed. Aliasing and phasing effects are obvious when viewing periodic targets such as those used for system characterization. Aliasing and phasing effects become pronounced as the target frequency approaches the electronic imaging systems's Nyquist frequency. Aliasing is not very obvious when viewing complex scenery and, as such, is rarely reported during actual system usage although it is always present. We have become accustomed to phasing effects and aliasing at the moves, on TV and on computer monitors. These effects becomes bothersome when trying to perform scientific measurements. What you see visually is not what you get with an EO imaging system.

Simulations of missile-ship-countermeasures engagements are used to determine the effective ways of defending a ship against infrared-guided missile threats. This paper describes one type of simulation that models the engagement of a ship deploying IR decoys by an infrared-guided seeker-head missile. This model was developed to assess the efficiency of IR decoys in protecting ships against these missiles. The simulation, Missile Infrared Decoy And Ship (MIDAS), is composed of three major blocks, the infrared scene generation, the seeker simulation and the missile dynamics simulation. The infrared scene generation block produces a three-dimensional IR scene from the target ship and flare models and transforms it into the two-dimensional IR image viewed by the seeker. The seeker simulation block is based on a generic conical scan seeker which uses a crossed-detector array for target detection. It processes the IR image to select a target and generates a steering command. The missile dynamics block computes the changes in missile trajectory according to the seeker steering command. The computations performed by each of the three blocks are explained in detail.

The significant reduction of the back scattering cross section of bounded layer composed of chaotically distributed resistive metallic film scatterers compared to the regular system of the same scatterers has been shown The theoretical estimates and the experimental measurements are in a good agreement and approach the experimental data for analogous bounded sand slab for nearly grazing angle of incidence.

In this paper we discuss the estimation of parameters describing polarized and partially polarized light when radiation is collected with a receiver consisting of an array of incoherent detectors and registering light levels after passage through an array of retardation plates and analyzers. The results obtained in semiclassical limit include Poisson photon statistics. We obtain fundamental limits on estimation of parameters of interest as a function of the number of photons available, and determine the value of a priori knowledge for achieving the desired estimation accuracy for both signal and background shot noise limited condition. The results presented are applicable to both active and passive systems where measurements of polarization signatures can improve detection and/or discrimination of targets, or when polarization coding is used to transmit information.

A multicomponent approach to calculate a light field structure with allowance for multiple scattering in the media such as clouds, mists and ocean water is given. All characteristic properties of the real phase function are regarded and propagation in a scattering medium of any optical thickness with an arbitrary single scattering albedo can be considered. The phase function is represented as a sum of more simple functions. The radiance is given as a sum of appropriate components. The equations like the radiative transfer equation are obtained for each component. They can be solved using the known methods within domains where they work best of all. Comparison of various solutions (lidar returns, temporal structure of light pulse transmitted by cloud, an angular structure of the light reflected from and transmitted by cloud) with different numerical calculations shows a fairly satisfactory accuracy of our analytical formulas.

A computer based thermal scene simulation was developed for generating synthetic images to facilitate the design and test of missile born Electro-Optical (EO) seeker systems. A scene is composed of objects and a background selected from the available data base files. Objects are modeled by combining small planar facets to cover their three dimensional surface. Each facet is associated in the data base with the thermal and optical properties of the surface being modeled. The objects may be stationary or mobile. The background is modeled on the interior surface of a pyramid defining the extent of the scene, with the floor of the pyramid representing the earth surface and the sides of the pyramid representing above horizon. The pyramid surface is then divided into triangular facets. Each facet is assigned a uniform texture selected from the database of background materials. The simulation is designed to be used with missile 6 degree of freedom (6DOF) simulations, and generates images as seen by the seeker based on the instantaneous seeker line of sight, seeker position, seeker orientation, object orientations and positions, and background.

Given the dependence of infrared imaging system performance on target thermal contrast, the accuracy in defining target (Delta) Ts is of critical importance. The current definition of (Delta) T neglects temperature variance across the target, and its inadequacy in characterizing infrared signatures is becoming increasingly apparent. With the advent of more sophisticated sensors and signal processors, with increased capabilities to extract more detailed thermal features, a new method of defining (Delta) T which includes temperature variance is required. This paper presents a new definition of thermal contrast, validates its improved capability to predict thermal imaging sensor performance with the available data, and addresses related issues including the effects of range and sensor resolution on the apparent temperature variance and the contribution of background clutter to the overall scene variance.

Two different solutions are given to the problem of estimation of time-varying target density functions in arbitrary nonstationary environments via use of one transmitted signal and a distributed sensor (or an array of point sensors). The first one utilizes a wideband signal with a certain bandlimitedness condition on the time variation of the target density function at each point in space. By choosing small successive processing time intervals, bandlimitedness of the time-variation becomes much more realistic than assuming no time-variation or a specific time-variation. The second one does not have this assumption and utilizes a narrowband signal with a continuous spectral content. These provide a general methodology for imaging of nonstationary environments. In addition to nonstationarity, time-varying density functions can also describe nonlinearity of a medium where each point distorts the transmitted wave by multiplying it with a time-varying function during reflection.

ERIM has developed a highly accurate method of simulating ocean backgrounds and targets in the infrared called Ship and Ocean Surface Image Simulation (SOSIS). This package provides realistic IR images of the ocean surface and horizon as a function of sea state, atmospheric conditions, sensor properties, and viewing geometry. SOSIS uses as inputs a thermal target model, an ocean surface model, and an environmental model. It samples these three models using a system of ray tracing and rough surface scattering to account for the interactions of the ocean surface, target, and the environment. At each ray-surface intersection, SOSIS takes into account small surface roughness, bi-directional reflection functions (BRDF's), solar glint, and polarization. The imagery from the SOSIS package is used for target signature prediction, weapons design, automatic target recognition (ATR) development, IR search and track (IRST) testing, ocean-horizon signatures, and many other functions.

A new concept is presented to determine the range of vision of objects in scattering media relied upon the idea of optimum image processing. A program package which realized this concept to calculate the limited range of detection and discrimination of underwater objects by any passive and active laser TV systems, including observation through a rough sea surface, is described. Some examples of vision range calculation when viewing in atmosphere and ocean using different systems from the human eye to pulsed range gating laser TV are given.

The Electro-Optics & Physical Sciences Laboratory of the Georgia Tech Research Institute (GTRI) continues to develop a physically-based scene simulation in support of a variety of applications including human factors studies and electro-optical system analyses. Accurate simulation of a scene requires the reproduction of the scene component shadows onto the simulated landscape. Scene simulation in the mid- and far-infrared wavebands provides a particularly difficult challenge due to the inherent long-term time dependence of the thermal interactions which define the heat flows in the scene. This paper will address an approach investigated at GTRI for implementing a shadowing algorithm within the GTVISIT scene simulation software. The paper will discuss the algorithms devised for computing signatures for shadow regions and subsequent rendering of those regions on the background. Example imagery illustrating the results of the algorithms will be shown.

The Battlefield Emission and Multiple Scattering (BEAMS) model computes the diffuse radiance fields in 3-dimensional clouds having spatially inhomogeneous concentrations and optical properties. The technique employs 26-stream radiative transfer calculations on a regular grid of aerosol concentrations. Incident illumination sources on the cloud include diffuse illumination, direct illumination, and finite width beam illumination. Volume emission effects as well as reflective and cyclic boundary conditions are also supported. The BEAMS model addresses a need for more realistic treatment of spatial fluctuations in aerosol and particulate clouds. Model comparisons are presented against known conditions, and examples are given of applications to modeling the spatial variability induced in background contrast and scene visualization.

Image statistics of measured and modeled cloud radiance are compared to evaluate the nature of radiance fluctuations. Images of aerosol and particulate clouds were obtained with the Atmospheric Transmission Large Area System (ATLAS). Radiative transfer model calculations were made for multiple scattering and emission in spatially inhomogeneous clouds of aerosols/particulates. Effects of spatial and time fluctuations, which are often neglected, are important because they may induce changes in background contrast and they affect the acceptability of simulated scenes.

In this paper we present an overview of research studies aimed at deriving quantitative measured for image clutter. Image clutter is identified as a perceptual effect and therefore quantitative measures for clutter should be derived on the basis of models for perception. Previous studies in the field have produced limited success since they concentrated on the signal level interpretation in clutter characterization. In this study, we promote examination of the clutter problem at a higher level, where preattentive cues form the basis for their definition. Specifically, we derive clutter measures based upon preattentive texture features. Results of an extensive experimental study highlight the limitations of signal level clutter measures and the utility of texture based measures.

Many of different descriptors of spatial properties of natural terrain and objects, in particular different texture descriptors, have been implemented. Using results from detection theory and image quality studies a set of texture measures has been selected by investigation of the amount of necessary uncorrelated measures. Using these we are able to measure the statistical multidimensional difference between terrain areas and object areas in a way that correlates with target acquisition performance.

An infrared all-sky mapper Scorpio has been developed at TNO-FEL. This system operates in the 3 - 5 micrometers and 8 - 13 micrometers wavelength bands. With its reflective optics system it scans a useful area of 340 degree(s) in azimuth and from -18 degree(s) up to +72 degree(s) in elevation with a resolution of 0.8 mrad. The system is calibrated during each revolution in an absolute way with the use of a temperature-controlled bar. The dynamic range is from -100 degree(s)C up to +300 degree(s)C. The mapper is used to obtain calibrated infrared images of the sky (primarily at 10 micrometers ) in an on-going study for statistical properties of the sky and cloud infrared radiation, i.e. infrared clutter and cloudiness. Scorpio is also useful for more general cloud and sky background studies. All-sky images are presented together with vertical profiles in empty and cloudy sky regions which are compared with Lowtran7 predictions of radiance, and meteorological cloud measurements. A description and analysis of the clutter analysis based on image cells is given. Images of RMS clutter are presented.

A high-frequency electromagnetic wave propagating through a randomly inhomogeneous medium will create random intensity pattern whose spatial structure is a function of the radiation wavelength and the parameters of the medium. The ability to detect the spatial variation and to find correlation between different patterns depends on the ratio between the receiver aperture and the characteristic spatial scale of the speckle. In this work using multiscale solutions for high-frequency random propagators we construct expressions for the integral intensity correlation measures which allow us to obtain spatial and spectral intensity correlation characteristics and to analyze their variation as functions of wavelength separation and detector aperture. Numerical results are presented for a plane wave propagating in a random medium characterized by a power law spectrum of the refractive index fluctuations.

Path integrated atmospheric transmittance over a 5.5 km horizontal path is measured using black target contrast. Measurements of on-line particulate distributions by Particulate Measuring System instrumentation (PMS) and of meteorological parameters are also made. The extinction coefficients, primarily scattering, of aerosols are calculated using the PMS data, and those arising from molecular absorption are calculated by LOWTRAN7. Both extinction coefficients, the directly measured path integrated and those calculated from particulate distribution and meteorological parameters near the receiver, are compared. Good agreement exists especially when relative humidity is low, despite the fact that the second method involves aerosol size distribution by data collected from a single point along the atmospheric path. Disagreement between both methods under high values of relative humidity can be explained by classification errors of the PMS instrumentation because of changes in the index of refraction of particles in a humid environment.

This paper proposes a methodology for parameterizing measured backgrounds using the magnitude and directionality of thermal fluctuations in a measured image. The proposed technique treats the background as a well behaved 2D random process with a regionalized power spectral density (PSD) defined by three parameters. A general framework for applying this PSD model to predict IIR sensor performance under the measured background conditions is presented. The framework uses the standard-deviation of the background in the vicinity of the target, which can be related to the PSD parameters depending on the target size. The background parameterization methodology was applied to a database of background images recorded as part of the TACAWS technology demonstration held in July 1992 at Redstone Arsenal, Alabama. The measurements were made in the 3-5 micrometers and 8-12 micrometers spectral bands using two imaging radiometers. Specific analysis of the parameterized background data is shown addressing the diurnal, spectral, and elevation-angle variations of the clutter parameters.

Ocean clutter in infrared images at low grazing angles is of interest for both airborne and shipborne detection of low flying targets over the ocean. In this paper a 2-D simulation model for infrared cutter at low grazing angle and data analysis results of three IRAMMP ocean tests are presented. The data analyzed correspond to IRAMMP calibrated dual band (mid-wave and long-wave) IR images. The observed ocean clutter is characterized by vertical profiles of the mean and variance, single channel power spectra, and the dual channel coherence as a function of environmental conditions and sensor geometry. The measured variance in radiance typically has a rapid increase with increasing look down angle below the horizon. Preliminary simulation predictions are consistent with data analysis results.

For the past 6 years, the Ft. Belvoir Meteorological Team has used the NOAA Wave Propagation Laboratory developed and Lockheed built path integrated optical scintillometer. This instrument has and continues to serve us well. Through research funding by the Army Test and Evaluation Command we were able to procure a new instrument for the measurement of C_n² which is based on holographic techniques. The 'HOLODAR' was developed by Dove Electronics, Inc. of Rome, N.Y. In this paper we describe the instrument and provide comparison data with the existing Lockheed system. It is our hope that HOLODAR can be used as a calibration reference for optical turbulence measurements.

The use of polarization as the principal parameter in target recognition and clutter reduction typically requires numerous field measurements with linear and circular polarizers in order to obtain a complete set of polarimetric images for a given target. We have developed a method which reduces the number of necessary field measurements to characterize the polarization state of a given scene. In polarimetric analysis with a single linear polarizer inserted in the light sensing path, it is found that three polarimetric images are sufficient to characterize the observed light field. These images are taken with the polarizer oriented at two mutually orthogonal directions and their bisecting direction respectively. These three polarimetric images form a basis set which allows the scene at any arbitrary polarization angle to be reconstructed. A synthetic polarimetric image data base can be easily realized with only these three images rather than the huge storage and retrieval requirements for a true measured data base. The coherency matrix of a quasi-monochromatic light wave together with Stokes vector and Mueller matrix are used to deduce the minimum required measurements in constructing a synthetic data base. Examples of measured and derived images will be compared and discussed.

Features of spatial filtering of partially coherent images are investigated. A radiooptical analogy method for analysis of optical system operation is used. Analytic expression of optical transfer function for arbitrary coherence of illumination field, for arbitrary aperture factors and for arbitrary defocusing parameter is deduced and investigated by numerical methods. Efficiency of spatial filtering of partially coherent images has been showed by laboratory experimental system using illumination source with variable coherence. An original method with two synchronously tunable moire patterns has been used for optical transfer function measurements. Investigation results are interpreted from the point of view of solving inverse problem.

The occurrence of background clutter is an on-going issue in the development of electro- optical sensors and seekers. Clutter varies with many parameters, which makes it costly to utilize measured data. Modeled backgrounds must be tested, however, to determine if the clutter they generate is realistic. The study reported here was performed with several goals: (1) to develop a methodology for studying clutter; (2) to compare the clutter levels from different scene elements; (3) study variations in spectral bandpass and in atmospheric visibility; and, (4) to study the effect of varying model sophistication on clutter. The last goal is one which has not previously been studied, to our knowledge. These results give model developers guidance on what model elements deserve the most resources. The present study focused on a generic reticle seeker, such as would be used in a tactical missile. The backgrounds studied were of tree-lines horizons, sun-heated rocks, and broken clouds, in four spectral bands within the 1 to 12 micron infrared region. Atmospheric haze levels were varied from 1 km to 23 km visibility. For these computations, the order of importance to clutter levels was, (1) scene type, (2) model sophistication level, and (3) haze. Strong variations with spectral band were also noted, although bands could not be compared fairly.

The focus of current research is the measurement, modeling, and simulation of the spatial, temporal and spectral behavior of gaseous radiation from a jet engine exhaust plume. Infrared imaging using a scanning infrared camera has been used as a flow visualization technique. Also, an infrared band model code has been developed to relate field variables to infrared emission. The model has been validated by comparisons with other infrared band models and by comparisons with experimental data. Finally, a free-jet apparatus has been constructed based on an auxiliary power unit, and a computer code has been developed for converting the experimentally measured exhaust plume temperature distribution to the corresponding infrared image. These efforts represent work in progress. The goals of this research are to better understand infrared imaging of exhaust plumes, to correctly model the infrared emission from exhaust plumes given knowledge of the temperature, pressure, and species distribution of the flow field, and finally to use this model to validate CFD codes through comparison of predicted and observed infrared images.

Work in progress at Georgia Tech to develop a model of human pattern perception, visual search, and detection is reviewed. The model's algorithms are based on research on low-level visual processes. Recent advances in the field have led to the development of computational models of the image processing performed by the visual system from the cornea to the striate cortex. The model also incorporates recent advances from research on visual search. The organization of the model for predicting target acquisition, analyzing target signatures, and specifying low-observable requirements is discussed.

Background images of a typical Mediterranean landscape were measured for 24 hours and their properties were analyzed. The locations were chosen to represent an area with various amount of vegetation. The thermal mean and variance changes around the clock were extracted from these background images and a typical pattern for the variance behavior was identified. A simple model is suggested for describing this behavior. The knowledge of variance dependence on the scene and measurement system parameters could be further used for simulation of IR images or for detection probability estimation.

The conventional area weighted average temperature (AWAT) (Delta) T is a primary performance measure for characterizing target/background scenes. However, the AWAT definition is widely recognized as being inadequate for representing observer sensitivity in many target detection and acquisition tasks. This situation is particularly true for targets which are at short ranges relative to the observer or viewed through powered optics. In these cases the mid and high spatial frequency components provide distinctive cue features which dominate over the average or aggregate characteristics of the target. The authors examine alternative definitions of (Delta) T in order to identify more robust and accurate metrics for the evaluation of sensor and signature countermeasure performance. The analysis indicates that target/background scene descriptions using simple average parameters such as the mean and standard deviation are not sufficient for characterizing imaging sensor performance against targets with internal texture and contrast gradients in background clutter.

Smart Weapons Operability Enhancement (SWOE) is a coordinated, cooperative Army, Navy, Marine Corps, and Air Force effort to integrate and validate a targets in backgrounds scene generation process. The SWOE Process has been developed to generate realistic multi- spectral, including IR and MMW, scenes representative of the greatest possible range of world wide battlefield scenarios. This Process is based on fundamental physics formulations which produce accurate temporal and spatial distributions of the energy fields reaching any given weapon sensor aperture. There are six areas of specific focus: (1) codification of methods to make measurements in support of synthetic generation of battlefield scenes, (2) automation of information bases construction for a world wide range of battlefields, (3) multi-spectral scene physics modeling for a world wide range of battlefield scenarios, (4) synthetic battlefield background environment and target rendering, (5) integration of synthetic battlefield scene generation capabilities onto common computer workstations, and (6) integration of the SWOE Process into an architecture that is compatible for use with state-of-the-art distributed networks.

The Smart Weapons Operability Enhancement (SWOE) Program has developed a package of databases, physics and scene rendering models, and analysis procedures that can be used by developers and testers of smart weapons to test and evaluate future smart weapons systems under realistic environmental conditions and natural backgrounds. This resource is being used to evaluate both infrared and millimeter wave systems. In the case of infrared systems, the driving force for the radiant fields that are ultimately sensed are the temperature fields in the scene. In the SWOE Program, a series of detailed thermal models are used to calculate the temperature conditions of the surfaces and objects in the scene. These models can be used to calculate the temperatures for a variety of surfaces and a full range of environmental conditions, including winter and non-winter conditions. This paper summarizes the one dimensional heat conduction model that is used to calculate the temperature conditions for surfaces without vegetation.

In this paper we outline the Smart Weapons Operability Enhancement (SWOE) cloud modeling requirements. We present an overview of the cloud model logic, and provide more detailed descriptions of the fractal field generation process and the cumuliform convection model.

This paper will detail the process of the rendering module of the Smart Weapons Operability Enhancement (SWOE) Process.

The paper presents a model for the detection of the noncoherent optical radiation, at the noise limit. Using the reprezentation on coherent states ofthe density matrix operators (p') and ofthe detection operators (iI), one can estimate: the detection probability (Qd ),the false-alarm probability (Qo), the signal-to-noise ratio (S/Z), as functions depending on the degrees of freedom of time oscillation modes (Mt) and space oscillation modes (Ms). Using Neyman-Pearson criterion, we build a validity algorhythm to validate the statistical hypotheses emerging from the analysis ofthe noncoherent optical field.

An image processing technique to ascertain target-to-background temperature differences of emissive sources of infrared (IR) radiation is discussed in this paper. After the processing system description, a discussion follows on the image (scene) acquisition process proceeded by an explanation of the image calibration procedure. An example image statistical analysis is presented to illustrate this technique. In particular, a temperature difference is derived for a U.S. M60 tank exercised in a desert background.

Calculating the temperature of surfaces with vegetation poses a particular challenge due to the extreme variability that can be encountered with vegetation. This paper summarizes how the Smart Weapons Operability Enhancement (SWOE) Program is modeling the temperatures for surfaces with vegetation. The types of vegetated surfaces that can be modeled include simple vegetation, such as grasses, forests of differing densities, and individual trees. The temperatures of the vegetated surfaces are calculated using a combination of one and three dimensional heat conduction models.

The primary objective of the Smart Weapons Operability Enhancement (SWOE) Joint Test and Evaluation (JT&E) Program is to validate the SWOE Process for the Office of the Secretary of Defense. The SWOE Process is a physics based scene generation capability that will enhance the performance of future smart weapon systems for a global variety of battlefield environments. This process is focused on generating complex background environmental scenes, including targets, for a world wide range of battlefield conditions. The SWOE program is a DoD wide partnership incorporating capabilities from the Army, Navy, Marine Corps and Air Force. The principal thrusts of SWOE are to quantitatively define the environmental factors and processes and to provide the capabilities to measure, model, render and extrapolate their impact on smart weapon system performance. The Grayling I exercise is the first in a series of four coordinate field deployments focused on validation of the SWOE Process. This paper describes the experimental design, sampling plans, measurement efforts and summarizes preliminary results of the Grayling I exercise.

Sets of image data from the MUSIC (Multi Spectral Infrared Camera) have been analyzed to obtain the spatial and temporal structure of cirrus and stratus cloud scenes at small spatial and temporal scales. The MUSIC data were collected in 1989 in a side-looking airborne geometry. The large clutter to noise ratios found in this analysis for both cirrus and stratus clouds represent a potentially large noise source for IRST. The power-law roll-off and anisotropy of the power spectral density of images is compared to the output of a 3-D cloud scene simulator and found to be in reasonable agreement; and in very poor agreement with simpler notions which invoke perspective distortion of two-dimensional clouds and predict much large anisotropy. The log of the time-lagged coherence of a registered sequence of images depends on the square of the spatial frequency and the square of the time lag. This is in agreement with a passive advection model of the cloud evolution, which successfully predicts the change in coherence behavior between smaller and large time lags relative to the advection integral time.

Target detection models are an important means of evaluating military systems and tactics. The models currently in use within the United States rely solely on first-order statistical measures of targets and backgrounds. Using a series of computer-generated images, this paper demonstrates that such measures ignore perceptually important information having a significant impact on detection of targets by personnel using unaided eyes, direct view optics, or electro-optical devices. Second-order statistical measures are proposed as a means of improving model performance. The paper provides a brief synopsis of current research tying second-order statistical measures to target detection models.

The authors report the statistical analysis of infrared scenes containing a military ground vehicle. The purpose is to attempt to determine the important variables in clutter as well as the robustness of the present definition of clutter through computer simulation. Both variance based and texture based clutter metrics are compared. The authors analyzed both cluttered and non-cluttered scenes.

EO/IR/Laser detection of a target amidst clutter/background is a difficult problem often treated with simplistic models. Unlike noise, clutter is more complex, neither spectrally white nor statistically Gaussian. Therefore, it is insufficient to lump clutter with noise and use standard detection curves. Battelle has produced image randomization software called BATRAN (Background and Target Randomization) which computes various types of statistical distributions to randomize background and target pixels separately. The types of statistics implemented include exponential, Gaussian, log-normal, and Rice distributions for both the background and target. In an effort to identify a more robust and accurate (Delta) T metric definition for background and target matching, Battelle also developed a new (Delta) T metric definition and its equation using RMS pixel-based higher order statistics for the background and target signature pixel data in a scene image. This new (Delta) T metric provides a better estimate of true signature difference between the background/clutter and target, enabling more accurate matching of the background/clutter and target for use in sensor detection performance assessment.