Threat Identification in Humanitarian Demining Using Machine Learning and Spectroscopic Metal Detection

Verre, Wouter van; Özdeğer, Toykan; Gupta, Ananya; Podd, Frank; Peyton, A. J.

doi:10.1007/978-3-030-33607-3_58

Cited by 7 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The advantages of using this approach to compute MPT coefficients over using commercial software, in terms of accuracy, computational efficiency and applicability to wide set of applications, has previously been discussed and demonstrated in References 3,4,6,12. With the goal of identifying hidden metallic targets in mind, the MPT spectral signature has been previously used for simple library classification, 13,14 a k nearest neighbors (KNN) classification algorithm, 15 and other machine learning approaches. 16 In addition, existing examples of practical MPT classification of objects include in airport security screening, 15,17 waste sorting, 18 and anti-personnel landmine detection. 19 In such situations, induced voltages are measured over a range of frequencies by a metal detector from which the MPT spectral signature of the hidden object is obtained and then a classifier applied.…”

Section: Introductionmentioning

confidence: 99%

“…With the goal of identifying hidden metallic targets in mind, the MPT spectral signature has been previously used for simple library classification, 13,14 a k nearest neighbors (KNN) classification algorithm, 15 and other machine learning approaches 16 . In addition, existing examples of practical MPT classification of objects include in airport security screening, 15,17 waste sorting, 18 and anti‐personnel landmine detection 19 .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identification of metallic objects using spectral magnetic polarizability tensor signatures: Object characterisation and invariants

Ledger

Wilson

Amad

et al. 2021

Numerical Meth Engineering

View full text Add to dashboard Cite

The early detection of terrorist threat objects, such as guns and knives, through improved metal detection, has the potential to reduce the number of attacks and improve public safety and security. To achieve this, there is considerable potential to use the fields applied and measured by a metal detector to discriminate between different shapes and different metals since, hidden within the field perturbation, is object characterisation information. The magnetic polarizability tensor (MPT) offers an economical characterisation of metallic objects that can be computed for different threat and non-threat objects and has an established theoretical background, which shows that the induced voltage is a function of the hidden object's MPT coefficients. In this article, we describe the additional characterisation information that measurements of the induced voltage over a range of frequencies offer compared with measurements at a single frequency. We call such object characterisations its MPT spectral signature. Then, we present a series of alternative rotational invariants for the purpose of classifying hidden objects using MPT spectral signatures. Finally, we include examples of computed MPT spectral signature characterisations of realistic threat and non-threat objects that can be used to train machine learning algorithms for classification purposes.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Identification of metallic objects using spectral magnetic polarizability tensor signatures: Object characterisation and invariants

Ledger

Wilson

Amad

et al. 2021

Numerical Meth Engineering

View full text Add to dashboard Cite

show abstract

“…The MPT spectral signature has been exploited in a range of different classification algorithms including simple library classification for homogeneous 15 and inhomogeneous objects, 16 a k nearest neighbors (KNN) classification algorithm 8 and other machine learning approaches. 17 The MPT classification of objects has already been applied to a range of different applications including airport security screening, 8,18 waste sorting, 13 and antipersonnel landmine detection. 10 The aforementioned supervised classification techniques rely on a library of MPT spectral signatures to learn how to classify the objects.…”

Section: Introductionmentioning

confidence: 99%

Efficient computation of the magnetic polarizabiltiy tensor spectral signature using proper orthogonal decomposition

Wilson

Ledger

2021

Numerical Meth Engineering

View full text Add to dashboard Cite

The identification of hidden conducting permeable objects from measurements of the perturbed magnetic field taken over a range of low frequencies is important in metal detection. Applications include identifying threat items in security screening at transport hubs, location of unexploded ordnance, and antipersonnel landmines in areas of former conflict, searching for items of archeological significance and recycling of valuable metals. The solution of the inverse problem, or more generally locating and classifying objects, has attracted considerable attention recently using polarizability tensors. The magnetic polarizability tensor (MPT) provides a characterization of a conducting permeable object using a small number of coefficients, has an explicit formula for the calculation of their coefficients, and a well understood frequency behavior, which we call its spectral signature. However, to compute such signatures, and build a library of them for object classification, requires the repeated solution of a transmission problem, which is typically accomplished approximately using a finite element discretization. To reduce the computational cost, we propose an efficient reduced order model (ROM) that further reduces the problem using a proper orthogonal decomposition for the rapid computation of MPT spectral signatures. Our ROM benefits from a posteriori error estimates of the accuracy of the predicted MPT coefficients with respect to those obtained with finite element solutions. These estimates can be computed cheaply during the online stage of the ROM allowing the ROM prediction to be certified. To further increase the efficiency of the computation of the MPT spectral signature, we provide scaling results, which enable an immediate calculation of the signature under changes in the object size or conductivity. We illustrate our approach by application to a range of homogenous and inhomogeneous conducting permeable objects.

show abstract

“…Measured MPT spectral signatures have previously been used in conjunction with a k nearest neighbours (KNN) classification algorithm [21] and other ML approaches [39]. In addition, existing examples of practical MPT classification of objects include in airport security screening [24,21], waste sorting [13] and anti-personal landmine detection [30].…”

Section: Introductionmentioning

confidence: 99%

Identification of Metallic Objects using Spectral Magnetic Polarizability Tensor Signatures: Object Classification

Wilson¹,

Ledger²,

Lionheart³

2021

Preprint

View full text Add to dashboard Cite

The early detection of terrorist threat objects, such as guns and knives, through improved metal detection, has the potential to reduce the number of attacks and improve public safety and security. To achieve this, there is considerable potential to use the fields applied and measured by a metal detector to discriminate between different shapes and different metals since, hidden within the field perturbation, is object characterisation information. The magnetic polarizability tensor (MPT) offers an economical characterisation of metallic objects and its spectral signature provides additional object characterisation information. The MPT spectral signature can be determined from measurements of the induced voltage over a range frequencies in a metal signature for a hidden object. With classification in mind, it can also be computed in advance for different threat and non-threat objects. In the article, we evaluate the performance of probabilistic and non-probabilistic machine learning algorithms, trained using a dictionary of computed MPT spectral signatures, to classify objects for metal detection. We discuss the importances of using appropriate features and selecting an appropriate algorithm depending on the classification problem being solved and we present numerical results for a range of practically motivated metal detection classification problems.

show abstract

Threat Identification in Humanitarian Demining Using Machine Learning and Spectroscopic Metal Detection

Cited by 7 publications

References 14 publications

Identification of metallic objects using spectral magnetic polarizability tensor signatures: Object characterisation and invariants

Identification of metallic objects using spectral magnetic polarizability tensor signatures: Object characterisation and invariants

Efficient computation of the magnetic polarizabiltiy tensor spectral signature using proper orthogonal decomposition

Identification of Metallic Objects using Spectral Magnetic Polarizability Tensor Signatures: Object Classification

Contact Info

Product

Resources

About