Landmines have been laid in conflicts around the world and continue to maim or kill civilians and soldiers.
Metal detectors (MD) have been used successfully to detect mines, but have difficulty detecting mines with little or no
metal content. Ground penetrating radar (GPR) systems have successfully been used to supply detection capabilities
where metal detectors fail. Handheld devices using such sensors have historically been used in battle but they can put the
user at high risk under direct fire from the enemy while exposed during some operations. We describe a robotic,
explosive hazard, anti-personnel/anti-tank mine detection system featuring dual-sensor GPR/MD capability for enhanced
mine detection and for removing the soldier from the mine field.
The MD is a broadband electromagnetic induction sensor to help discriminate between buried landmines and
metal clutter. The sensor operates in the frequency domain and collects data at 21 logarithmically spaced frequencies
from 300 Hz to 90 kHz. The GPR is a broadband stepped frequency continuous wave (SFCW) sensor operating from
700 MHz to 4 GHz in 10 MHz steps. The GPR employs an array of low cross section inverted V-dipoles swept over the
scene. The GPR data will also support 3D synthetic aperture radar (SAR) imagery to aid in user target verification.
The legacy AN/PSS-14 (Army-Navy Portable Special Search-14) Handheld Mine Detecting Set (also called
HSTAMIDS for Handheld Standoff Mine Detection System) has proven itself over the last 7 years as the state-of-the-art
in land mine detection, both for the US Army and for Humanitarian Demining groups. Its dual GPR (Ground Penetrating
Radar) and MD (Metal Detection) sensor has provided receiver operating characteristic curves (probability of detection
or Pd versus false alarm rate or FAR) that routinely set the mark for such devices. Since its inception and type-classification
in 2003 as the US (United States) Army standard, the desire for use of the AN/PSS-14 against alternate
threats - such as bulk explosives - has recently become paramount. To this end, L-3 CyTerra has developed and tested
bulk explosive detection and discrimination algorithms using only the Stepped Frequency Continuous Wave (SFCW)
Ground Penetrating Radar (GPR) portion of the system, versus the fused version that is used to optimally detect land
mines. Performance of the new bulk explosive algorithm against representative zero-metal bulk explosive target and
clutter emplacements is depicted, with the utility to the operator also described.
Proc. SPIE. 6553, Detection and Remediation Technologies for Mines and Minelike Targets XII
KEYWORDS: Detection and tracking algorithms, Scattering, Sensors, Feature extraction, Signal processing, Mining, Algorithm development, Electromagnetic coupling, Land mines, General packet radio service
The Region Processing Algorithm (RPA) has been developed by the Office of the Army Humanitarian Demining Research and Development (HD R&D) Program as part of improvements for the AN/PSS-14. The effort was a collaboration between the HD R&D Program, L-3 Communication CyTerra Corporation, University of Florida, Duke University and University of Missouri. RPA has been integrated into and implemented in a real-time AN/PSS-14. The subject unit was used to collect data and tested for its performance at three Army test sites within the United States of America. This paper describes the status of the technology and its recent test results.
CyTerra's dual sensor HSTAMIDS system has demonstrated promising landmine detection capabilities in extensive government-run field tests. Further optimization of the successful PentAD algorithm is desirable to maintain the high probability of detection (Pd) while lowering the false alarm rate (FAR). PentAD contains several input parameters, making such optimization using standard Monte-Carlo techniques too computationally intensive. Genetic algorithm techniques, which formerly provided substantial improvement in the detection performance of the metal detector sensor algorithm alone, have been applied to further optimize the numerical values of the dual-sensor algorithm parameters in more practical time frames. Genetic algorithm techniques have also been applied to choose among several sub-models and fusion techniques to potentially train the HSTAMIDS system in new ways. An analysis of genetic algorithm results has indicated that ground type may have a significant impact on the optimal parameter set. In this presentation we discuss the performance of the resulting ground-type based genetic algorithm as applied to field data.
Proc. SPIE. 5794, Detection and Remediation Technologies for Mines and Minelike Targets X
KEYWORDS: Radar, Principal component analysis, Standoff detection, Detection and tracking algorithms, Calibration, Metals, Data corrections, Ground penetrating radar, Land mines, General packet radio service
The Handheld Standoff Mine Detection System (HSTAMIDS system) has achieved outstanding performance in government-run field tests due to its use of anomaly detection using principal component analysis (PCA) on the return of ground penetrating radar (GPR) coupled with metal detection. Indications of nonlinearities and asymmetries in Humanitarian Demining (HD) data point to modifications to the current PCA algorithm that might prove beneficial. Asymmetries in the distribution of PCA projections of field data have been quantified in Humanitarian Demining (HD) data. The data suggest a logarithmic correction to the data. Such a correction has been applied and has improved the FAR on this data set. The increase in performance is comparable to the increase shown using the simpler asymmetric rescaling method.
Proc. SPIE. 5415, Detection and Remediation Technologies for Mines and Minelike Targets IX
KEYWORDS: Radar, Principal component analysis, Standoff detection, Detection and tracking algorithms, Calibration, Metals, Automatic target recognition, Ground penetrating radar, Land mines, General packet radio service
Outstanding landmine detection has been achieved by the Handheld Standoff Mine Detection System (HSTAMIDS system) in government-run field tests. The use of anomaly detection using principal component analysis (PCA) on the return of ground penetrating radar (GPR) coupled with metal detection is the key to the success of the HSTAMIDS-like system algorithms. Indications of nonlinearities and asymmetries in Humanitarian Demining (HD) data point to modifications to the current PCA algorithm that might prove beneficial. Asymmetries in the distribution of PCA projections of field data have been quantified in Humanitarian Demining (HD) data. An initial correction for the observed asymmetries has improved the False Alarm Rate (FAR) on this data.
CyTerra's dual sensor HSTAMIDS system has demonstrated exceptional landmine detection capabilities in extensive government-run field tests. Further optimization of the highly successful PentAD-class algorithms for Humanitarian Demining (HD) use (to enhance detection (Pd) and to lower the false alarm rate (FAR)) may be possible. PentAD contains several input parameters, making such optimization computationally intensive. Genetic algorithm techniques, which formerly provided substantial improvement in the detection performance of the metal detector sensor algorithm alone, have been applied to optimize the numerical values of the dual-sensor
algorithm parameters. Genetic algorithm techniques have also been applied to choose among several sub-models and fusion techniques to potentially train the HSTAMIDS HD system in new ways. In this presentation we discuss the performance of the resulting algorithm as applied to field data.
This paper answers in the affirmative the question: will it ever be feasible to predict useful infrared buried mine detection performance? The infrared (IR) is essentially blind at certain hours, but can have excellent vision at other times. The trick to making the IR a tactically useful tool is to plan mine detection operations during its best time of utility. Rather than use thermal models with their difficulty in representing IR imagery, we used a matched filter detector on IR video, in combination with prediction techniques using neural nets and weather data, to show that weather conditions can be successful in predicting IR mine detection performance. Prediction using mine detection models and weather data, correlated using neural nets and then predicted using weather data alone is not only theoretically feasible, but is also practical. Feasibility was demonstrated in Train A/Test A mode, where the neural nets achieved 100% prediction accuracy for both AP and AT mines. Practicality was demonstrated using single day Train A/Test B results, where 98% to 88% accuracy was achieved for AT mines from 2.5 to 12.5 hours forward, respectively. The technique is expected to be limited only by the accuracy of the short-term weather forecast.
A multi-sensor approach to buried object discrimination has been developed by Coleman Research Corporation (CRC) as a practical successor to currently prevalent metal detectors. The CRC multi-sensor unit integrates with and complements standard metal detectors to enable the detection of low- metallic and non-metallic anti-tank and anti-personnel mines as well as the older metallic-jacketed mines. The added sensors include Ground Penetration Radar (GPR) and Infrared (IR). The GPR consists of a lightweight (less than 1 LB) snap on antenna unit, a belt attached electronics unit (less than 5 LB) and batteries. The IR consists of a lightweight (less than 3 LB) head mounted camera, a heads-up virtual display, and a belt attached processing unit (Figure 1.1). The output from Automatic Target Recognition algorithms provide the detection of metallic and non-metallic mines in real-time on the IR display and as an audio alert from the GPR and MD.
The maturation and commercialization of uncooled focal plane arrays and high density electronics now enables lightweight, low cost, small camera packages that can be integrated with hard hats and military helmets. It is only recently that low weight, staring long wavelength infrared (LWIR) sensors have become available employing uncooled focal planes at array size and sensitivities that provide enough information for useful, man-portable, wearable applications. By placing the IR camera on the head, a hands-free infrared virtual reality is presented to the user. This paper describes applications, the design of a helmet mounted IR sensor and presents images from the helmetcam. The head gear described has a noise equivalent delta temperature (NEDT) of less than 50 milliKelvin, consumes less than 10 watts and weighs less than 3 kilograms.
Software neural nets hosted on a parallel processor can analyze input from an IR imager to evaluate the likelihood of a buried object. However, it is only recently that low weight, staring LWIR sensors have become available in uncooled formats at sensitivities that provide enough information for useful man-portable helmet mounted applications. The images from the IR are presented to a human user through a see-through display after processing and highlighting by a neural net housed in a fanny-pack. This paper describes the phenomenology of buried object detection in the infrared, the neural net based image processing, the helmet mounted IR sensor and the ergonomics of mounting a sensor to head gear. The maturing and commercialization of uncooled focal plane arrays and high density electronics enables lightweight, low cost, small camera packages that can be integrated with hard hats and military helmets. The head gear described has a noise equivalent delta temperature (NEDT) of less than 50 milliKelvin, consumes less than 10 watts and weighs about 1.5 kilograms.
Infrared imagery scenes change continuously with environmental conditions. Strategic targets embedded in them are often difficult to be identified with the naked eye. An IR sensor-based mine detector must include Automatic Target Recognition (ATR) to detect and extract land mines from IR scenes. In the course of the ATR development process, mine signature data were collected using a commercial 8-12 (mu) spectral range FLIR, model Inframetrics 445L, and a commercial 3-5 (mu) starting focal planar array FLIR, model Infracam. These sensors were customized to the required field-of-view for short range operation. These baseline data were then input into a specialized parallel processor on which the mine detection algorithm is developed and trained. The ATR is feature-based and consists of several subprocesses to progress from raw input IR imagery to a neural network classifier for final nomination of the targets. Initially, image enhancement is used to remove noise and sensor artifact. Three preprocessing techniques, namely model-based segmentation, multi-element prescreener, and geon detector are then applied to extract specific features of the targets and to reject all objects that do not resemble mines. Finally, to further reduce the false alarm rate, the extracted features are presented to the neural network classifier. Depending on the operational circumstances, one of three neural network techniques will be adopted; back propagation, supervised real-time learning, or unsupervised real-time learning. The Close Range IR Mine Detection System is an Army program currently being experimentally developed to be demonstrated in the Army's Advanced Technology Demonstration in FY95. The ATR resulting from this program will be integrated in the 21st Century Land Warrior program in which the mine avoidance capability is its primary interest.
Alternative algorithms for detecting and classifying mines and minelike objects must be evaluated against common image sets to assess performance. The Khoros CantataTM environment provides a standard interface that is both powerful and user friendly. It provides the image algorithmist with an object oriented graphical programming interface (GPI. A Khoros user can import 'toolboxes' of specialized image processing primitives for development of high order algorithms. When Khoros is coupled with a high speed single instruction multiple data (SIMD) algorithms. When Khoros is coupled with a high speed single instruction multiple (SIMD) processor, that operates as a co-processor to a Unix workstation, multiple algorithms and images can be rapidly analyzed at high speeds. The Khoros system and toolboxes with SIMD extensions permit rapid description of the algorithm and allow display and evaluation of the intermediate results. The SIMD toolbox extensions mirror the original serial processor's code results with a SIMD drop in replacement routine which is highly accelerated. This allows an algorithmist to develop identical programs/workspace which run on the host workstation without the use of SIMD coprocessor, but of course with a severe speed performance lost. Since a majority of mine detection componenets are extremely 'CPU intensive', it becomes impractical to process a large number of video frames without SIMD assistance. Development of additional SIMD primitives for customized user toolboxes has been greatly simplified in recent years with the advancement of higher order languages for SIMD processors (e.g.: C + +, Ada). The results is a tool that should greatly enhance the scientific productivity of the mine detection community.
We are building automatic target recognizer (ATR) systems. These systems are being applied to many different target detection scenarios. Our work has been in the military application field, but the problems are the same for most commercial applications as well. The measures of performance are the same. How well can a human perform the same target detection task? What is the probability of detecting (Pd) the target versus the false alarm rate (FAR)? The community has evolved comparative performance techniques that present the merits of alternative system approaches. In this paper, we present the results of a comparative study of alternative algorithms for detecting and classifying buried and surface land mines from an airborne platform in infrared imagery. The results show that for low signal-to-clutter ratios, more complex algorithms produce higher Pd for a given FAR. More complex algorithms signify the need for a high performance, high throughput processor to meet typical time lines. An update on the geometric arithmetic parallel processor (GAPPTM) high performance/throughput machine is therefore provided.
Land mine detection and extraction from infra-red (IR) scenes using real-time parallel processing is of significant interest to ground based infantry. The mine detection algorithms consist of several sub-processes to progress from raw input IR imagery to feature based mine nominations. Image enhancement is first applied; this consists of noise and sensor artifact removal. Edge grouping is used to determine the boundary of the objects. The generalized Hough Transform tuned to the land mine signature acts as a model based matched nomination filter. Once the object is found, the model is used to guide the labeling of each pixel as background, object, or object boundary. Using these labels to identify object regions, feature primitives are extracted in a high speed parallel processor. A feature based screener then compares each object's feature primitives to acceptable values and rejects all objects that do not resemble mines. This operation greatly reduces the number of objects that must be passed from a real-time parallel processor to the classifier. We will discuss details of this model- based approach, including results from actual IR field test imagery.
The focus of this paper is on the classifier module for a real-time automatic target recognition system. It is a dynamic neural network that utilizes adaptive, on-line learning. The on-line learning component is required because of the changeability of the target detection scenario resulting in unpredictable feature space representation of targets and clutter. The system was successfully tested against IR imagery of target models.
This paper describes the work completed by Martin Marietta in support of the U.S. Army's standoff minefield detection system, advanced technology transition demonstration. This paper discusses the high priority and urgent need for the standoff mine detection system within the Army Combat Engineers, it presents the results of the successful application of non developmental technology/hardware in an airborne mine/minefield detection system, and it discusses the significant payoff of applying advanced ATR and high speed parallel processing. The technologies discussed include the IR imager as the source of mine imagery, advanced image processing algorithms including neural nets, and a high speed parallel processor unique to Martin Marietta called GAPP (geometric arithmetic parallel processor).