Characterising buried objects in metal detection

TL;DR

This paper develops a new mathematical model for detecting highly conducting buried objects in metal detection, accounting for soil properties like conductivity and permeability, leading to improved identification accuracy.

Contribution

It introduces an asymptotic expansion incorporating ground conditions and a magnetic polarizability tensor for better object characterization.

Findings

Enhanced model accuracy over traditional methods

Tensor coefficients can be computed independently of soil properties

Improved object identification in varied ground conditions

Abstract

Current mathematical models for identifying highly conducting buried objects in metal detection assume that the soil is non-conducting and has the same permeability as free space. However, although the electrical conductivity of soil is low, it is not negligible and depends on factors such as soil type and salinity. Moreover, the magnetic permeability of soil varies with its iron content and is often described as uncooperative. Depending on the ground conditions, these soil's properties can influence the induced voltages in the measurement coils of metal detectors and becomes of increasing importance as the frequency of the exciting current source is increased. In this work, we develop a new asymptotic expansion for the perturbed magnetic field due to the presence of a highly conducting magnetic buried object as its size tends to zero, which takes account of the ground conditions. The leading order term of this expansion can be expressed in terms of a complex symmetric rank-2 magnetic polarizability tensor, which characterises the object, can assist in its identification, and we provide conditions under which this tensor's coefficients can be computed independently of the ground conditions. We demonstrate the improved accuracy of our new result, which takes account of the ground conditions, over the situation where the soil's conductivity and permeability are not considered.