UXO detection using small loop TEM method

Transient Electromagnetic (TEM) and Magnetic Resonance Sounding (MRS) methods were applied jointly to investigate lithological variations in the Dead Sea coast of Israel. The subsurface in this area is heterogeneous and composed of intercalated sand and clay layers over a salt rock, which is partly karstified. Groundwater is very saline, with a chloride concentration of 100-225 g/l. TEM is known as an efficient tool for investigating electrically conductive targets like saline water, but it is sensitive to the salinity of groundwater, the clay content and the porosity of rocks. MRS, however, is sensitive primary to groundwater volume but also to lithological variations in the subsurface. MRS is much less sensitive to variations in groundwater salinity in comparison with TEM. We show that MRS enables us to resolve the fundamental uncertainty in TEM interpretation caused by the equivalence between groundwater resistivity and lithology. Combining TEM and MRS in Nahal Hever area we have identified lateral extensions of different lithological formations.

Liquid water can exist at temperatures well below freezing beneath glaciers and ice sheets, where subglacial water systems, fresh and saline, have been shown to host unique microbial ecosystems. Geophysical techniques sensitive to fluid-content contrasts, e.g. electromagnetics, can characterize subglacial water and its salinity. Here, we assess the ground-based transient electromagnetic (TEM) method for deriving the resistivity and salinity of subglacial water. We adapt an existing open-source Bayesian inversion algorithm, which uses independent depth constraints, to output posterior distributions of resistivity and pore fluid salinity with depth. A variety of synthetic models, including a thin (5 m), conductive (0.16 Ωm), hypersaline (147 psu) subglacial lake, are used to evaluate the TEM method for imaging under 800 m-thick ice. The study demonstrates that TEM methods can resolve conductive, saline bodies accurately using external depth constraints, for example, from radar or seismic data. The depth resolution of TEM can be limited beneath deep (>800 m), thick (>50 m) conductive, water bodies and additional constraints from passive electromagnetic (EM) methods could be used to reduce ambiguities in the TEM results. Subsequently, non-invasive active and passive EM methods could provide profound insights into remote aqueous systems under glaciers and ice sheets.

The removal of native vegetation for the development of land has caused the salinisation of surface soil and water resources in Australia. It is important to determine the existence of potential salinity hazards so that preventative measures may be taken at an early stage. The current hydrogeological conditions and the topography of a given catchment influence the movement of saline groundwater, and it is of particular importance to identify areas of high- and low-recharge to groundwater. On a catchment scale, the delineation of high-recharge areas by drilling can be prohibitively expensive, and more efficient mapping methods are required. Broadband electromagnetic (EM) sounding methods offer the possibility of obtaining the vertical conductivity distribution in the regolith, to produce three-dimensional maps of the resistivity structure. These maps could be used to interpret the distribution of features related to hydrology, such as recharge and stratification of stored salt in catchments. For salinity-related problems, a regolith depth of about 50 m is relevant. It is therefore necessary to measure the high-frequency content of the EM response to differentiate shallow conductive layers. For both ground-based and airborne EM methods, the response has been modelled in a frequency range of a few hundred Hz to 50 kHz. Salinity profile types have been identified to characterise different types of recharge. These profile types have been simulated with models consisting of horizontal layers of different resistivity. The theoretical modelling has determined that EM methods can be used to distinguish different types of salinity profiles. These results are supported by inversion of EM data collected in the field in a number of catchments. The transient electromagnetic (TEM) method is used for ground-based broadband EM measurements, and modelling shows that both the depth and thickness of the salt accumulation can be resolved when its depth is as little as 6 m and its thickness 10 m. Additionally, a relatively resistive basement below the salt accumulation may be distinguished from a conductive one. Since the former case implies the presence of a high-recharge salinity profile and the latter case a low-recharge profile, this method may be used to differentiate the two types of profiles. Down-hole conductivity logs in catchments near Collie and at East Yornaning, WA, confirm that different salinity profiles can be distinguished with EM methods. At Collie, the TEM method has detected the presence of salt accumulations and has determined the depth of their lower boundary. Inversion of these results shows the presence of a resistive electrical basement below the salt accumulations, and therefore indicates that relatively high recharge conditions are associated with these salinity profiles. Similarly, measurements along traverse lines in the East Yornaning catchment show a correlation between the inversions results and known geological features.

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.

The airborne transient electromagnetic (TEM) method is widely used in land applications but faces challenges in marine applications due to the strong masking effects of seawater. The accuracy of the inverse Laplace transform algorithm strongly affects the applicability of the 2.5D TEM method in marine research; thus, an appropriate transmitted waveform is required. To resolve these issues, a trapezoidal wave was utilized as the waveform of the current source, owing to the substantial energy contribution of the low-frequency range. Further, the Talbot algorithm was selected for the inverse Laplace transform as it can yield an accurate response with smaller summation terms than the commonly used Gaver–Stehfest (G-S) algorithm. On this basis, a rugged seabed and a subsea tunnel were also investigated. The voltage response is obtained when the flight heights of the loop source change. The results confirmed that the Talbot algorithm with a trapezoidal waveform is more reliable and robust for complex marine geological models and is expected to provide an effective approach for marine explorations.

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.

The ground‐airborne transient electromagnetic method with a galvanic source has such advantages as high efficiency, great exploration depth, high signal to noise ratio and good performance in mountainous areas. However, so far its interpretation system has not been established, which restricts the development of this method. This paper aims at building up a complete ground‐airborne electromagnetic detection system to enrich the theory of this method. This paper consists of three parts: (1) full field apparent resistivity definition; (2) Kirchhoff immigration of TEM virtual wave field; and (3) inverse synthetic aperture imaging. Magnetic field intensity is selected for full field apparent resistivity definition. The results of theoretical models show that the first and last part of the apparent resistivity curve can reach the resistivity of the first and the last layer, respectively. Thus, full space and full time apparent resistivity definition is achieved. Based on previous study, Kirchhoff migration of ground‐airborne transient electromagnetic method with a long wire source is realized. A correlation method is adopted for multiple survey point synthesis, which pushes the traditional single point processing forward point‐by‐point for multi‐coverage inverse synthetic aperture imaging.The results of layered models show that full field apparent resistivity can reveal the underground electric change smoothly, completely and gradually. When the resistivity of the middle layer changes, the differentiation of full field apparent resistivity is obvious, which can easily and intuitively identify the electric layer. As receiver altitude, offset and time are considered simultaneously, full field apparent resistivity can be calculated in full space and full time. The results of the complex model with a water‐bearing gob shows that the further the offset from the gob is, the weaker the resistivity anomaly on full field apparent resistivity section is. There are two interfaces from the Kirchhoff immigration results, of which the upper one indicates surface. As the electrical difference between earth and air is large, the wave field signal of this interface is strong, which spreads over the whole area. While the anomalies of the lower interface are mainly distributed in the middle area and weaken outwards, which indicate the water bearing gob. The upper interface changes little after inverse synthetic aperture imaging, while the gob anomaly narrows and has a clear boundary. The anomaly boundary matches well with the designed model. Inverse synthetic aperture imaging of ground‐airborne transient electromagnetic survey with a long wire source is realized on the basis of inverse synthetic aperture radar (ISAR). Full field apparent resistivity is put forward with inverse function and iteration, which can realize calculation of this variable in the full zone and full time. Pseudo‐seismic migration of transient electromagnetism is used in 3D transient field imaging. Inverse synthetic aperture radar is introduced in inverse synthetic aperture imaging of the ground‐airborne transient electromagnetic method with a long wire source, which can help enhance resolution. The results of designed models show that correlation can strengthen useful signal and improve signal to noise ratio and resolution, which prove the efficiency of inverse synthetic aperture imaging of the ground‐airborne transient electromagnetic method with a long wire source.

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%.

Joint use of TEM and MRS methods in a complex geological setting

Transient electromagnetic method (TEM) is regarded as an everlasting issue for geological exploration. It is widely used in many research fields, such as mineral exploration, hydrogeology survey, engineering exploration and unexploded ordnance detection. The traditional measurement systems are often based on ARM、DSP or FPGA, which have not real-time display, data preprocessing and data playback functions. In order to overcome the defects, a real-time data acquisition and preprocessing system based on LabVIEW virtual instrument development platform is proposed in the paper, moreover, a calibration model is established for TEM system based on a conductivity loop. The test results demonstrated that the system can complete real-time data acquisition and system calibration. For Transmit-Loop-Receive (TLR) response, the correlation coefficient between the measured results and the calculated results is 0.987. The measured results are basically consistent with the calculated results. Through the late inversion process for TLR, the signal of underground conductor was obtained. In the complex test environment, abnormal values usually exist in the measured data. In order to solve this problem, the judgment and revision algorithm of abnormal values is proposed in the paper. The test results proved that the proposed algorithm can effectively eliminate serious disturbance signals from the measured transient electromagnetic data.

In recent years, several traditional mining exploration methods have been successfully modified for use in the environmental geophysics field. In these cases, "successfully modified" refers primarily to acquiring data fast enough, and therefore economically efficient enough, to accommodate the relatively small budgets that are available in most environmental studies.For example, transient electromagnetics (TEM) methods have been increasingly applied to environmental problems, particularly in unexploded ordnance (UXO), underground storage tank (UST) and utilities detection. Small, in-loop, cart-mounted TEM measurements are excellent for these deep metal detection applications. A major research effort is now underway to use, among other methods, multi-component, multi-time-gate mobile TEM systems (measuring Hx and Hy, as well as the standard Hz) in order to discriminate targets of interest (UXO, for example) from non-UXO objects such as metallic debris.A second good example is the induced polarization (IP) method. Although resistivity has been used extensively in shallow environmental applications, IP data acquisition has always been too slow, and therefore too expensive, for most environmental targets. Multichannel receivers, multiplexers, and laptop computers now allow us to acquire IP data at rates of 2500 to 3000 data points per day (in the dipole-dipole configuration, for example), providing low cost, high density data. IP data have been shown to be particularly useful in delineating buried waste, such as at old landfills. In numerous surveys, resistivity data have provided useful information on buried, back-filled pits and trenches, while IP data have been used to discriminate which of the pits and trenches located with resistivity actually contain waste material.

The transient electromagnetic (TEM) method is a geophysical exploration technique used in a wide range of applications such as mineral exploration, hydrogeology surveys, and underground geological mapping. In recent years, it has been applied in tunnel construction to forecast geological anomalies in front of the excavation face. However, the TEM response can be affected by metallic objects such as excavation carts and metallic support structures, which makes interpretation more difficult. The purpose of this study is twofold: to introduce a new method for extracting and eliminating TEM interference from metallic bodies in a tunnel excavation environment and to gain insight regarding the influence of metallic objects on the TEM response by physical modeling in the laboratory and studies conducted in the field. The concept of resistive limits is introduced; these limits can be used as eigenvalues in the inversion process, thus achieving more accurate identification of both the location and size of ball-shaped, low-resistivity anomalies (gouge materials) in surrounding rock that has high resistivity. Both the physical laboratory simulations and the field experiments provide insight on the interference effect of metal objects such as a steel arches or excavation carts on the TEM response. Field experiments inside a tunnel indicate that the maximum distance between the excavation cart tested and the tunnel face that can influence the TEM response is 30 m, while the maximum influence distance of a steel arch is 1 m in the vertical plane of the coil and 2 m in the axial direction. The proposed inversion process was shown to be capable of extracting the interference of metallic objects from TEM responses and improve the accuracy of TEM interpretation in identifying geological anomalies.

Unexploded ordnances (UXOs) are a big problem in this country and in the world. They are often buried in unknown locations, and therefore need to be found before they can be excavated. Currently, a major focus of UXO related research is reducing the large number of false positive detections to each true positive detection. This paper proposes a new step to the current detection and localization strategy. This addition develops a suboptimal trajectory that will maximize the amount of information returned when a vehicle revisits a suspected UXO location and reexamines the area. The resulting adaptive sensing technique utilizes either the sigma-point Kalman filter or the extended information filter to quantify the estimated information obtained in each sensor run. The algorithm is first tested with a range sensor and then implemented with a magnetometer. The results show improvements in UXO detection. The sigma-point Kalman filter approach produces a more accurate solution while the extended information filter algorithm runs faster.

Transient electromagnetic (TEM) method is a reliable means of non‐destructive testing for fault diagnosis of grounding grid in substation. In view of the small transmitter loop of the TEM system and many measuring points and survey lines of grounding grid detection in substation. This study presents a fast solution to TEM apparent resistivity imaging for grounding grid detection using artificial neural networks. The input ‐output mapping relation of neural network is established based on TEM response characteristics of the grounding grid. The built network could map the recorded TEM data of grounding grid detection and quickly obtain the resistivity image. The proposed method offers accuracy and fast computation for the resistivity imaging of grounding grid. Feasibility and technical attractiveness of the proposed method in fast imaging of apparent resistivity is investigated with the measurement of grounding grids in the actual substations. It can be used in real time so that the recorded TEM data in substation can be calculated without re‐training, which avoids time‐consuming inversion computation. The rapid processing of massive data and submitting the detection results of the grounding grid to customers rapidly and in real time, which is expectation for a modern power industry.