Adaptive Sensing for Improving Detection of Unexploded Ordnances

A few promising solutions for thermal imaging Unexploded Ordnance (UXO) detection were proposed after the start of the military conflict in Ukraine in 2014. At the same time, most of the landmine clearance protocols and practices are based on old, 20th-century technologies. More than 60 countries worldwide are still affected by explosive remnants of war, and new areas are contaminated almost every day. To date, no automated solutions exist for surface UXO detection by using thermal imaging. One of the reasons is also that there are no publicly available data. This research bridges both gaps by introducing an automated UXO detection method, and by publishing thermal imaging data. During a project in Bosnia and Herzegovina in 2019, an organisation, Norwegian People’s Aid, collected data about unexploded ordnances and made them available for this research. Thermal images with a size of 720 × 480 pixels were collected by using an Unmanned Aerial Vehicle at a height of 3 m, thus achieving a very small Ground Sampling Distance (GSD). One of the goals of our research was also to verify if the explosive war remnants’ detection accuracy could be improved further by using Convolutional Neural Networks (CNN). We have experimented with various existing modern CNN architectures for object identification, whereat the YOLOv5 model was selected as the most promising for retraining. An eleven-class object detection problem was solved primarily in this study. Our data were annotated semi-manually. Five versions of the YOLOv5 model, fine-tuned with a grid-search, were trained end-to-end on randomly selected 640 training and 80 validation images from our dataset. The trained models were verified on the remaining 88 images from our dataset. Objects from each of the eleven classes were identified with more than 90% probability, whereat the Mean Average Precision (mAP) at a 0.5 threshold was 99.5%, and the mAP at thresholds from 0.5 to 0.95 was 87.0% up to 90.5%, depending on the model’s complexity. Our results are comparable to the state-of-the-art, whereat these object detection methods have been tested on other similar small datasets with thermal images. Our study is one of the few in the field of Automated UXO detection by using thermal images, and the first that solves the problem of identifying more than one class of objects. On the other hand, publicly available thermal images with a relatively small GSD will enable and stimulate the development of new detection algorithms, where our method and results can serve as a baseline. Only really accurate automatic UXO detection solutions will help to solve one of the least explored worldwide life-threatening problems.

Unexploded ordnance (UXO) investigation is mainly carried out by the magnetic exploration. However, in this paper, the transient electromagnetic (TEM) method was used in order to detect UXO more accurately. Although the TEM method is usually used for deep sounding of mineral exploration, we used small loop TEM method to detect buried UXO. In order to understand the characteristics of the small loop TEM method, many experiments using small empty cans were carried out at Ito Campus in Kyushu University and a lot of features were discovered. For example, if the size of loop becomes smaller, the penetration depth becomes smaller. These results showed that the small loop TEM was applicable to UXO investigation. Based on these results, the test survey to detect the imitation UXOs using 180×180-cm-square loop was carried out. In this field experiment, there are various kinds of imitation UXOs, but all of the UXOs were detected. As a result, it was found that UXO detection by small loop TEM method is effective compared to magnetic method. Moreover, the small loop TEM method could narrow the range of UXO location to the size of 30×30cm-square.

This paper summarizes the development of a wireless magnetoresistive (MR) sensor array for unexploded ordnance (UXO) detection realized with wireless sensor network (WSN) technology. In recent years, the rapid development and research on WSN has enabled the application of such technology in a variety of areas¿DThis paper discusses the WSN implementation in a UXO and landmine detection system. The overall system design, hardware design, and network design for large area field survey for a passive UXO detection is discussed. The manufactured wireless MR sensor node prototype is extremely compact, lightweight and portable. It also demonstrated excellent overall system performance in house testing. A pilot production of the sensor nodes is proposed for large scale field testing to further verify the effectiveness of the wireless MR sensor array for UXO and landmine detection.

Magnetic mapping for the detection and characterization of UXO: Use of multi-sensor fluxgate 3-axis magnetometers and methods of interpretation

The fundamental problem in applying geophysical mapping to locate unexploded ordnance (UXO) is distinguishing true UXO from non-UXO. Enhancing the accuracy of UXO detection has multiple benefits, especially in the areas of cost savings and safety. We investigated discrimination approaches using both magnetic field data and numerically modeled data. Libraries of total field magnetic (TFM) responses were calculated using finite element modeling for three UXO types found at a Montana National Guard training site. UXO model parameters were varied over ranges of azimuth, declination, and depth resulting in approximately 600 models per UXO type. The modeled responses of finite-element model (FEM) and actual TFM field data were then used as training data in discrimination and classification approaches comparing neural networks (NN), random forests (RF), and support vector machines (SVMs). The prediction targets in the training process comprised three classes: 1) binary [UXO or noninteresting object (NIO)]; 2) multiclass (UXO round type and NIO); and 3) classes derived from multiclass self-organizing feature map (SOFM) analysis. The multiclass SOFM targets generated from site-specific field data were found to be optimal for UXO discrimination. The best performing combination of class selection types using recentered data for UXO detection rates of 100% resulted in a false alarm rate (FAR) of 28%.

A team of US Army Corps of Engineers, Omaha District and Engineering and Support Center, Huntsville, let Propulsion Laboratory (JPL), Stanford Research Institute (SRI), and Montgomery Watson is currently in the process of planning and conducting the largest ever survey at the Former Buckley Field (60,000 acres), in Colorado, by using SRI airborne, ground penetrating, Synthetic Aperture Radar (SAR). The purpose of this survey is the detection of surface and subsurface Unexploded Ordnance (UXO) and in a broader sense the site characterization for identification of contaminated as well as clear areas. In preparation for such a large-scale survey, JPL has been developing advanced algorithms and a high-performance restbed for processing of massive amount of expected SAR data from this site. Two key requirements of this project are the accuracy (in terms of UXO detection) and speed of SAR data processing. The first key feature of this testbed is a large degree of automation and a minimum degree of the need for human perception in the processing to achieve an acceptable processing rate of several hundred acres per day. For accurate UXO detection, novel algorithms have been developed and implemented. These algorithms analyze dual polarized (HH and VV) SAR data. They are based on the correlation of HH and VV SAR data and involve a rather large set of parameters for accurate detection of UXO. For each specific site, this set of parameters can be optimized by using ground truth data (i.e., known surface and subsurface UXOs). In this paper, we discuss these algorithms and their successful application for detection of surface and subsurface anti-tank mines by using a data set from Yuma proving Ground, A7, acquired by SRI SAR.

Buried unexploded ordnance (UXO) and land mines are grave threats to civilians who go back to contaminated areas in the postwar period. In this research, we propose a new technique to detect UXOs from ground penetrating radar (GPR) signals. The technique is based on an energy-based feature of the region of interest (ROI). A ROI selection technique in range domain is also proposed. The proposed method was tested with the UXO data collected by Battelle company and the Ohio State University. This data set was made available to public through the Unexploded Ordnance Center of Excellence (UXOCOE), Department of Defense, U.S.A. The results evaluated in terms of Receiver Operating Characteristic (ROC) curves suggest that our proposed technique performs very well. Comparisons to a traditional detection technique and our previously proposed technique are conducted. The improvement by the proposed technique does not only speed up the UXO/land mine clearance operation, but also save operating cost.

Over the past five years, our group has focused on development of airborne magnetic and electromagnetic systems for mapping and detection of unexploded ordnance (UXO). Among these projects is one involving testing and development of a high temperature airborne Superconducting Quantum Interference Device (SQUID) tensor magnetometer. In association with the airborne SQUID development, we are conducting tests of the SQUID as a ground‐based instrument for UXO mapping and detection. The ground‐based SQUID is mounted on a man‐portable litter with ancillary instrumentation mounted on a cart and connected by an umbilical. High‐temperature SQUIDs use liquid nitrogen as the cryogen, which is more convenient than the helium‐cooled low‐temperature devices. Preliminary data from the ground‐SQUID prototype show good agreement with cesium vapor total magnetic field data along a test line. Noise frequencies and levels appear manageable and we anticipate that specialized processing should make it possible to extract more details about UXO items from their tensor magnetic properties.

Sub-Audio Magnetics (SAM) is a method by which a total-field magnetic sensor may be used to simultaneously acquire both total magnetic intensity (TMI) and total-field electromagnetic induction (TFEMI) responses from buried metal. In situations where public safety is involved, regulators, such as the U.S. Environmental Protection Agency, can mandate the use of both magnetic and electromagnetic sensors for the detection of potentially hazardous items such as unexploded ordnance (UXO) and metal drums. SAM has now completed a program of evaluation and certification as a state-of-the-art technology, acceptable for application in the litigation-sensitive environmental hazard remediation market.During its certification program, SAM was applied to controlled U.S. test sites at the Aberdeen Proving Ground, Maryland, McKinley Range, Alabama, Waikoloa, Hawaii, and at the Newholme facility in Australia. It was also applied to situations in Montana where live UXO was present. In these demonstrations, a direct comparison was made between SAM results and common, state-of-the-art alternative geophysical detecting technologies. Based on demonstrations at these sites, it has been possible to quantify the key performance parameters: level of effort, detection performance as a function of depth and target size, false-alarm rejection achieved through dual-mode sensing, and the cost-benefit ratio of SAM compared with contemporary sensor technologies such as the Geonics EM61 and EM63 electromagnetic detectors, and G-tek’s TM-5EMU electromagnetic detector and TM-4 total field magnetometer.SAM has proven to be very successful in detecting a wide range of metallic UXO items to depths exceeding achievable detection by all contemporary sensors. In the Montana live sites, where the smallest item of UXO was a 76 mm projectile, false-alarm rates were reduced from 7 per UXO, using TMI only, to 0.7 per UXO using combined SAM TMI and TFEMI data, without diminishing the probability of UXO detection. Survey cost per hectare using SAM was determined to be less than one quarter that of conducting TMI plus EMI surveys separately.The application of SAM to extremely magnetic lava flows at Waikoloa in Hawaii showed that SAM cannot yet compete with the ground-balancing capability of the TM-5EMU in this environment. Advanced signal-processing development may reduce this present limitation.

: This report the findings and recommendations of a study, conducted for the Environmental Security Technology Certification Program (ESTCP) to identify potential barriers to the transfer of technologies from the research and development phase to active field use for munitions detection. To conduct a practical, focused assessment of barriers to technology transfer, the study focused on two specific emerging technologies: Wide Area Assessment (WAA) and unexploded ordnance (UXO) discrimination. This report discusses WAA; a parallel report on UXO discrimination is also available. The issue of UXO detection and remediation technology transfer barriers has been discussed for the last decade, perhaps most notably in the Final Report of the Defense Science Board Task Force on Unexploded Ordnance1. That report identified two changes that would bring about significant cost savings: use WAA to reduce the footprint of potential Munition Response Sites (MRSs) and use discrimination technologies to reduce the number of holes dug looking for munitions. Over the lifecycle of an environmental action starting with site investigation (SI) and working through remedial investigation/feasibility study (RI/FS) and remedial action (RA) towards site closure, WAA has applicability on the front end and UXO discrimination has applicability during the RA phases (see Figure 1). The need for cost-effective WAA methods and technologies is driven by the magnitude of the effort the Department of Defense (DoD) must undertake to clean up UXO on former and active DoD lands. The DoD must examine millions of acres of land potentially contaminated with UXO to identify areas that actually contain UXO for cleanup. On much of the acreage perhaps 80% or more there are no UXO present; this area is presumptively clean and should be accepted as needing No Further Action (NFA) by all stakeholders.

In this paper, we propose a two-stage model for unconstrained face detection. The first stage is based on the normalized pixel difference (NPD) method, and the second stage uses the deformable part model (DPM) method. The NPD method applied to in the wild image datasets outputs the unbalanced ratio of false positive to false negative face detection when the main goal is to achieve minimal false negative face detection. In this case, false positive face detection is typically an order of magnitude higher. The result of the NPD-based detector is forwarded to the DPM-based detector in order to reduce the number of false positive detections. In this paper, we compare the results obtained by the NPD and DPM methods on the one hand, and the proposed two-stage model on the other. The preliminary experimental results on the Annotated Faces in the Wild (AFW) and the Face Detection Dataset and Benchmark (FDDB) show that the two-stage model significantly reduces false positive detections while simultaneously the number of false negative detections is increased by only a few.

9 - The contribution of geophysics to landmine and unexploded ordnance (UXO) detection: Case study in the Egyptian environment

The task of detecting and locating landmines, unexploded ordnances (UXO) and improvised explosive devices (IEDs) is still a major challenge up to the present day. Problems such as the distance between the hazardous area and the measurement system as well as the differentiation between the target of interest and the surrounding soil are of importance in the development of the sensor system. Various types of radar based systems have been developed over the last decades to solve these problems. Compared to other methods ground penetrating synthetic aperture radar (SAR) has the ability to scan large areas from a safe standoff distance in a relatively short time. In this paper, two different imaging radar systems are compared. The first one is a ground-based SAR (GB-SAR) developed at German Aerospace Center (DLR). The other system is an unmanned aerial vehicle-based SAR (UAV-SAR) from the University of Ulm. The presented data originates from a joint campaign using the same measurement scenarios.

The detection and classification of unexploded ordnance (UXO) is considered a multidimensional pattern recognition problem. Standard techniques in solving multidimensional detection and classification problems involve using large sets of templates or libraries. This paper shows that by using wavelet transformation a single library will allow a particular class of ordnance to be classified over a range of depths.

Ground penetrating radar (GPR) has become at present an instrument widely used in different domains. This paper is based on the using of GPR for detection and characterization of unexploded ordnances (UXO). The obtained experimental data and some signal and image processing procedures are presented in order to make a correct evaluation of type of UXOs, caliber, dimensions and their location in the horizontally plan as well in depth.