Rapid, real-time reconnaissance of a large ground area is critical for the commander on the battlefield. It provides him with the ability to gather valuable knowledge of troop and armor movement, and tactical intelligence in any direction in a matter of minutes instead of days. A passive infrared line scanner which covers a large area in a short time will provide this effective tactical reconnaissance in the battlefield. The Airborne Minefield Detection and Reconnaissance System (AMIDARS) has been built by CAI, a Division of RECON/OPTICAL, INC., under contract to the CECOM Center for Night Vision and Electro-Optics, to meet this need. The AMIDARS Infrared Line Scan-ner (IRLS) provides three-axis stabilized, real-time, day/night imagery with a 120° Field of View (FOV). This paper discusses the essential features of the AMIDARS sensor design.

This paper describes Northrop's developmental Chukar 111-R reconnaissance system, which is based on the Chukar III target drone. Some military needs for reconnaissance and the advantages of employing an unmanned air vehicle to satisfy these needs are noted. Next, features incorporated into the new Chukar III-R reconnaissance system are described. These features include a high performance unmanned air vehicle (UAV), an infrared line scanner for imaOng targets, radio position-fix enhanced navigation, and a new mission planning and control station. Sensor slight test results, a pay-load mockup, and mission planning and image exploitation capabilities are discussed. The advantages of high speed and low observability are cited. Launch and retrieval techniques are described.

The design criteria for framing thermal imagers are well documented. However those for thermal linescanners are less easily available & are normally directed towards sensors optimised for fast, low flying manned aircraft. This paper aims to outline the considerations & characteristics of an infra-red linescanner specifically designed & optimised for UMA ( Un-Manned Aircraft ) or RPV (Remote Piloted Vehicle ) use. One of the major problems that has to be overcome is the lack of reliable real time interactive operator control so that the system must be capable of efficient autonomous operation under all operational conditions. The design of this system is described and in particular the control of the thermal gain of the scanner is considered. Here, as the scene is not multiply scanned, the use of the standard iterative algorithms derived for framing sensors could cause a serious performance degradation causing targets to be missed. To solve this problem the characteristics of various target scenes were analysed, (these included land, sea and coastal targets, both in the clear and in clutter). A method was derived for the linescanner to measure scene characteristics and to implement a predictive gain control to maximise target information. Also covered is the design of the main optical assembly to minimise manual alignment and simplify any field stripping that may be required and to maximise the reliability of a precise optical system in the highly hostile RPV environment. Finally the performance of the Barr And Stroud Infra-red Linescanner ( BASIL ), built from the above design work , is summarized together with the operational experience gained.

This paper will present a modular digital Recce Management System that provides for a large number of user configurations, depending on sensor suits, on-board or ground recording and requirements of airborne evaluation. The paper will concentrate on a description of the functions with deeper analysis of features that characterize the Ericsson Recce Management System as a system that is adaptable to a wide array of future reconnaissance demands.

Manned reconnaissance flights have been driven down to the lowest possible altitudes. Usually airborne optical sensors have less coverage when they are used at extremly low altitudes. Therefore an additional oblique viewing capability is mandatory. Infrared linescanners used during the past 20 years are restricted in their field of view, as well as in resolution at low altitudes and high speeds. Since scan angles of only 120 degrees had been achieved, the resolution presently reached at 200 ft altitude was good enough to identify the target or even to analyze the imagery.

Selection of payload components of a modern day remotely piloted vehicle (RPV) or unmanned air vehicle (UAV) is complicated by many factors. The requirement for real time and near real time data from an air vehicle of limited size, and therefore payload capacity, is a difficult prospect. This paper explores the latest electro-optic (E0) systems available to meet the mission requirements as well as the limitations of such devices and their associated systems. The RPV/UAV payload consists of sensor, recorder/playback and data link components. Resolution is a primary consideration in mission effectiveness. Resolution drives system bandwidth and conversely bandwidth limitations restrict system resolution.

As sensor payload for the CL-289 drone system, Carl Zeiss developed a new compact reconnaissance camera system. The high performance camera with 143 degree wide angle ground coverage, is a pulse operated sequential frame camera. It features true-angle foreward motion com pensation across the entire format and direct stereoscopic viewing. Small size and low weight permits easy installation in remotely piloted vehicles, pods and aircraft.

Honeywell EOD's linescanner has been enhanced for real-time operation by downsizing the AN/AAD-5 and repartitioning the design for adaptability to a variety of applications. The repartitioning effort has developed several new line replaceable units that allow real-time data linking and display of linescanner data. The near real-time E-0 design allows in flight recording and simultaneous downlinking on a video bandwidth data link. A simple, lightweight ground station real-time display has also been developed that can process the data linked imagery. This display will record, display, and process the imagery. This paper will discuss the architec-ture of this downsized E-0 system. The new real-time sensor has been flight tested and a production run for a foreign customer is nearing completion.

In this paper we discuss the work being done at Itek combining parallel, symbolic, and neural methodologies at different stages of processing for imagery exploitation. We describe a prototype system we have been implementing combining real-time parallel image processing on an 8-stage parallel image-processing engine (PIPE) computer with expert system software such as our Multi-Sensor Exploitation Assistant system on the Symbolics LISP machine and with neural computations on the PIPE and on its host IBM AT for target recognition and change detection applications. We also provide a summary of basic neural concepts, and show the commonality between neural nets and related mathematics, artificial intelligence, and traditional image processing concepts. This provides us with numerous choices for the implementation of constraint satisfaction, transformational invariance, inference and representational mechanisms, and software lifecycle engineering methodologies in the different computational layers. Our future work may include optical processing as well, for a real-time capability complementing the PIPE's.

As new high-detailed imaging systems develop, new keys to interpretation must also be developed, especially when distance and perspective are changed dramatically. Low altitude, large-scale reconnaissance (LALSR) imagery is close enough to the ground to eliminate many of the traditional visual cues that exist in most aerial images. The perspective involved can be compared to looking at rust on wire bristles without seeing the whole brush. Now, individual species of plants are recorded, rather than general types of vegetation. Because of increased surface detail with LALSR, a publication covering image interpretation (II) keys would be too voluminous. For this reason, a user approach to the development of II keys for LALSR appears to be more useful and economical. The user, knowing his needs and environment, can best select and create those keys which apply. The selected keys will be based on the visual cues, perspective, and scales, peculiar to LALSR. Keys of this type can be enhanced by the system operator during flights because of his proximity in the target area. The potential of this system for on-site processing also assists the user in key definition. Example projects are outlined.

Herein is the description of the methodology we adopted to develop a set of algorithms performing the automatic recognition and localisation of sites which are observed through an IR camera from a flying mobile. Considered sites are solid buildings such as houses, power-stations... They must be significant enough to allow satisfactory recognition. However they may include planar subparts like roads, greenfields,... To achieve this recognition, 3D site models are recomputed from CAD models to which are added selected attributes. Chosen models are sets of polyhedral facets which may be processed as derived sets of vertices or edges as well. Polyhedral models are particularly fitting general infrared image properties. Geometrical information is worked from the very beginning of the segmentation process. Image processing procedures extract visual features fitting at best the selected model constituents. At first, a 2D image graph is backprojected into a 3D graph thanks to the model (prediction) and then projection onto the 2D space carries the verification from the generated 3D hypotheses, until matching and localisation are completed. Sporadic monocular images are supposed to be output from an infrared camera. Nevertheless radar images, when available, are concurrently supplied. Provided simple data fusion process, radar information improves greatly the detection of emerging sites and the focus of attention on limited areas of the infrared image, from which the effective recognition is performed. A first implementation of the system is currently under completion relying on edge-based models. Extended use of models allowing feature cooperation is planned and other features like points of interest, regions are already taken into account.

The suppression of a wildfire is analogous to a combat action. Fires, like battles, spread fast and suppression forces must be highly mobile. The enemy, (in this case) the wildfire, is lethal in that it kills or destroys forces, equipment, and natural resources left in its path. The suppression action must be carried on day and night until the "enemy" is contained. Both air operations and ground forces are used. Just as in a combat situation, wildfire suppression forces need penetrating reconnaissance with real time data transfer. This paper presents a review of the current system of intelligence gathering on a wildfire where aerial observers, infrared detectors, and ground intelligence officers gather data and either radio or carry the data to the command center. It then attempts to show how some current military reconnaissance systems might be applied to wildfire control processes. The payoffs would include improved safety for both air and ground forces and faster containment of the wildfire which would reduce forest resources lost and decrease the total monetary cost of the containment action.

In the commercial arena, there is a variety of off-the-shelf hardware and software available to provide low cost, easily transportable multiispectral imagery (MSI) processing systems. Taking advantage of available equipment, and recognizing military interest in MSI, the Department of Defense (DoD) initiated the Multispectral Imagery Testbed (MIT) Project to assess the utility of a com-mercial off-the-shelf (COTS) MSI processing system for military applications. Geodynamics Corporation custom-designed the MIT using off-the-shelf hardware and software. The MIT is built around a COMPAQ 386/20 microcomputer configured with ERDAS high resolution image processing hardware and software and a variety of input/output devices. The MIT was deployed to four locations for 10-week periods, to include 2 weeks of hands-on training and 8 weeks of operational support. This paper describes the methodology used for system design and source selection and the results of the deployments.

CAI, a Division of RECON/OPTICAL, INC. is currently developing a High-Resolution Display System (HRDS) capable of displaying real-time imagery from the Airborne Minefield Detection and Reconnaissance System (AMIDARS) data stream or a tape unit. This paper provides a system overview, modes of operation, a discussion of the system configuration and current development status.

The interpretability of imagery is a subjective judgement drawn by experienced analysts. Their use of context and previous experience permits them to draw remarkable conclusions from photographic evidence. Specifications for the ATARS electro-optical (E0) sensors are written to demand performance defined on the Image Interpretability Rating Scale (IIRS). The rating achieved is based on the interpreter's evaluation of the useful information content compared to examples that define the scale. Examples at each level are associated with a typical range of ground resolvable distances. Hardware is designed to deliver an angular resolution corresponding to the desired ground sampled distance at the worst case altitude. This paper addresses the issue of predicting sensor performance by relating the ground sampled distance, which is an objective design parameter, to the ground resolvable distance, which is not.

A comprehensive model for calculating the degrading effects of atmospheric turbulence on the performance of long-range, airborne imaging systems is described. The model output is the Modulation Transfer Function (MTF) of the turbulent air in the conical ray bundle from the target to the sensor. Published data from the field of astronomy and other theoretical developments are incorported into the formulation. Two difficulties with previous formulations are avoided: the requirement to set a parameter to one of two values for near or far field geometry; and the requirement to choose between two different formulations for short and long exposures. These are handled in an automatic and continuous manner in this new model. Also, it was found that the index of refraction turbulence structure coefficient (Cn2) altitude profile has about the same value and appearance over most of the world; therefore, the Cn2 altitude profile is integrated and a plot of the results (parameter Ch) is included.

Images of an airborne, scanning, radiometer operating at a frequency of 98 GHz, have been analyzed. The mm-wave images were obtained in 1985/1986 using the JPL mm-wave imaging sensor. The goal of this study was to enhance the information content of these images and make their interpretation easier for human analysis. In this paper, a visual interpretative approach was used for information extraction from the images. This included application of nonlinear transform techniques for noise reduction and for color, contrast and edge enhancement. Results of the techniques on selected mm-wave images are presented.

The original MACE (Minimum Average Correlation Energy ) filters are addressed with attention to the effect of iterative refinements, new database tests (on strategic relocatable objects, missile launch-ers), depression angle and resolution effects on the number of training set imagery required and noise performance. Major attention is given to our new MACE filter algorithms for distortion-invariant pattern recognition: iterative shift filter synthesis, MV-MACE (Minimum Variance-MACE) filters (for improved noise performance), multiple symbolic encoded filters and Gaussian Minimum Mean Square Error(MMSE) filters.

The evolution of large (488 x 512 pixel) platinum silicide area imager detectors has made conceptually viable similarly large buttable TDI (Time Delay and Integration) detectors for use in wide area coverage airborne recon-naissance. We treat here with the evolution of a sensor "rear end" design which makes practical the development of an efficient and operationally flexible reconnaissance camera for use in the 3-5 micrometer spectral "window".

Today's changing tactical environment emphasises the NATO forces need for a common, real-time data link to provide the dissemination of intelligence data. This paper addresses the NATO problem, but the requirements and the scenario are applicable for other theatres as well. The NATO theatre was chosen because it represents the worst case situation in terms of the threat environment and the density of emitters. It is imperative that common linkage among forces is available and that the networks for the dissemination of critical data are in place and survivable; i.e., the node points must be away from prime target areas. The purposes herein are threefold. First, a review of some of the desirable aspects of a data link, as envisaged by the operational community is presented. Second, a NATO Recce scenario with its attendant data collection requirement is postulated. Third, pertinent data link characteristics, for operation in this environment, are suggested. Data link technology limitations, trade-offs and representative configurations are presented, with an emphasis on interoperability issues. Although no universal solutions are presented, it is shown that a NATO interoperable data link is feasible.

The requirements of military commanders for near real-time images of battlefield situations has long been frustrated F photographic film-based reconnaissance systems that need time consuming chemical processing. This has led to ti development of magnetic tape-based reconnaissance systems in which the aerial camera has been replaced with sophisticate electro-optical (EO) sensors. In the EO system, the electrical signal output from the EO sensor is digitized and recorded on tap while the reconnaissance aircraft is making a high-speed, low level pass over the battlefield targets to be imaged. Upon recut to the recovery site, just minutes after imaging the targets, the aircraft transmits the recorded images over a microwave dat link to an analysis facility at or near the battlefront. The tactical information thus received is only moments old and can be actel upon immediately. The high data rates involved in recording and reproducing the high-resolution images required by the El reconnaissance systems has necessarily imposed new and demanding performance requirements on the magnetic tape recorde These performance requirements have been met by the DCTR-A120 airborne recorder designed and built by DATATAP Incorporated.

The ability to send photographic information to command centers is a vital element in performing effective near real-time reconnaissance and surveillance operations. This imagery, in conjunction with other battlefield data, provides the battlefield commander with up-to-date intelligence for making decisions. Until recently, the ability to provide this real-time information was severely restricted by the logistics of physically moving, developing, and then disseminating the film. This time delay resulted in out-of-date, stale intelligence. This problem situation has eased recently due to technological developments that have been instrumental in facilitating the dissemination of near real-time information to forward operating areas and behind enemy lines. The Naval Air Development Center (NAVAIRDEVCEN) has capitalized on these developments and established the Tactical Imaging System (TIS). This miniaturized, man-pack, SATCOM/HF transmitting system provides near real-time tactical imagery. It consists of an image sensor, image intensifier, zoom lens, and image transmission processor. This paper provides an overview of the TIS components, specifications, operations, and future developments and applications. The TIS will have potential application in areas such as identification (friend or foe), reconnaissance, surveillance, and battlefield assessment. Under the TIS program, NAVAIRDEVCEN has developed hands-on experience in still video images and related technologies, including fleet satellite communications, HF transmission, image compression algorithms, VSLI integrated circuitry design, and day/night imagery techniques. NAVAIRDEVCEN has developed a complete, miniaturized system to conduct operational demonstrations, and to demonstrate operational tactics and utilization concepts. This paper provides an overview of the TIS components, specifications, operations, and future developments and applications.

New data recorder technologies are now available to help cope with the collection, storage and processing of digitized picture information. Given the massive quantity of data and rates at which such data must be transferred, rotary scanning type recorders and improved I/O processing are required. These new recorders must capable of operation in hostile environments, yet have the operational flexibility to support both data collection and subsequent analysis.

Autonomous lunar and planetary landers require a sensor for real time hazard detection during the terminal descent phase when landing in rough, hazardous terrain. The sensor must establish surface feature data from which on-board data processors can identify safe touchdown spots and provide guidance commands for the descent engines. The sensor should increase the probability of safe touchdown(no landing failure) from the less than 90% estimated for the Mars Viking mission to the current NASA planning requirement of ..98% for the Mars Rover Sample Return mission. The sensor must also satisfy mass, volume and power constraints imposed by the limited lander resources. We analyzed the sensor performance required, assuming the desired surface feature data is a range versus angle map of the accessible landing zone. Existing active and passive sensor technologies were evaluated for the estimated requirements of spatial resolution(lm at 2 km), working slant range(2 km), range precision(0.25m) and frame rates ?_ 1Hz. Pixel mapping rate was identified as the critical sensor parameter because of limited descent time and lander power. We have developed a concept of an active scanned laser radar and scan algorithm for a hazard sensor that meets the requirements using available technology. The sensor concept and supporting system analysis and trades will be presented.

This paper surveys rotary head; helical scan, digital recorders available or under development for recon-naissance applications. It begins with a brief review of the recorder parameters used in categorizing and comparing recorders. Data rate and record time, two of the more important parameters, because they help determine the record/reproduce technology required, are then used to define three general catego-ries. Commercial TV recorder development progress is used to project what is likely to become available in the near future. Lastly, these categories are compared for data rate overlap and other characteristics.

As the reconnaissance community transitions from film to near real time electro-optics, one obstacle has been the data rate limitations of complementary hardware. A modern linear array sensor can easily produce more bits of imaging data than data links, recorders or hard copy devices can handle. This problem is compounded by TDI and staring arrays with still higher pixel counts. While array and sensor manufacturers continue to strive for resolution equivalent to film, the manufacturers of the complementary hardware are also addressing the steps necessary for this transition. However, this task is made even more difficult due to volumetric constraints and cost concerns. This paper describes a small format recorder (less than 19mm) based on Super VHS or S-VHS technology. Rather than being a modified commercial unit, the Super SVCR V-301-A has been designed specifically for the severe environment encountered in airborne applications. Details of the recorder's development are presented, and engineering confidence and environmental testing are also summarized. Current and future growth applications, especially those relating to reconnaissance, are also provided.

Performance requirements for Long Range Oblique Photography (LOROP) cameras are a principal driver for system stabilization requirements. Systems analysis can predict the stabilization level required for a given Ground Resolved Distance (GRD) under specific conditions. This information is used then as a basis for the stabilization design in which choices for bearings, passive isolators and active devices can be selected. In this paper, we show examples of GRD's versus residual rate errors for various ranges of a panning system. We also discuss how both active and passive isolation techniques can be used to achieve a desired performance level. The CA-990 LOROP camera system is used as an example of how multifaceted stabilization techniques can be used to create a very precise pann-ing camera system in a high-vibration environment.

The target collimator test set for the tube-launched, optically tracked, wire-guided (TOW) missile launch system is reviewed and tested to facilitate calibration with emphasis on system alignment. Test set objective lens dispersion data and Gaussian analysis for visible and infrared (IR) wavelengths are included; associated optical point spread and modulation transfer curves obtained from observations at 0.6328 um with a Wyko model 6000 interferometer are also exhibited. In addition, test set bandpass filter transmittance spectra in the visible and near IR, obtained with a Bomem model DA3.02 Fourier-transform interferometer spectrometer, are displayed as well as mean transmittance values weighted with standard photopic data over indicated spectral intervals. Evaluation of the data presented indicates the need for more stringent requirements for test set bandpass filter specifications in order to reduce the recently emerged visual alignment problem essentially caused by chromatic aberration correlated with use of new stock filters. Primarily due to this result, a new design version of the test set employing a reflective collimator would be more appropriate to cover the required visible, 575 nm, and near IR, 0.94 um and 1.25 um peak nominal wavelengths, provided it is sufficiently cost-effective to implement. This investigation should be of interest and useful for those engaged in related optical alignment work especially for those involved with electro-optic calibration support of TOW.

Manned and unmanned tactical airborne reconnaissance systems of the future will use common mission data management functions in both airborne and ground support equipments to perform similar missions while meeting the requirement for commonality and interoperability. Fairchild Communications & Electronics Company's Mission Support System (MSS II) and Data Transfer Equipment (DTE) are developed to meet these functional needs. Application of off-the-shelf hardware to perform the sometimes different manned and unmanned mission data management functions are discussed.

Long Range Oblique Photographic (LOROP) cameras must provide quality images at long ranges despite distur bances from the carrying aircraft. A method of reducing these disturbances is termed image stabilization, and usually. consists of both passive and active mechanisms. Two-axis gimballing of a pointing mirror for the camera has proven. to be adequate to provide three-axis image stabilization. This paper describes the technique and the equations govern-, in.g the stabilization.

The transportable Digital Enhancement Database (DEDS) System stores and processes image data in a fully self-contained system. It accepts imagery from a variety of sources such as video tape, video maps, or digital/CCD camera. It includes generalized and customized image processing and manipulation functions. It also provides for special processing to enhance imagery from selected sensors. Imagery can be annotated and stored on magnetic disk (on line) or on optical disk for archiving, storage, and retrieval. Data relating to this imagery is stored in a database and is accessible for update or review. The database drives the selection of imagery for retrieval. A scheme for selecting images based upon desired characteristics of the frames and subsequent selection of specific images visually from a "visual index" has been implemented. A Mapping Subsystem is included in the DEDS. A wide variety of maps are available on an analog video disk and can be retrieved, annotated, and stored with embedded annotation along with other acquired imagery or within its own subsystem. The system has been designed with growth as a major parameter. Incorporation into much large scale systems such as those for real-time mission planning and control is envisioned.

The RMS-1000 Reconfigurable Mobile System is a highly mobile ground station with direct application to the needs of users requiring real-time access to information gathered by airborne reconnaissance platforms. The system combines high-performance computing and image processing workstations with command, control, and communications capabilities in a lightweight self-contained shelter. The RMS-1000 provides a capability for receiving and processing many forms of intelligence data, including imagery. Background information is presented which shows the RMS-1000 as the culmination of over 38 years of tactical ground station development. The motivation for development of the RN/IS-1000 is further discussed in the context of military requirements in a European conflict.