Object Identification Using Multifrequency Emi Data

Abstract

Unexploded ordnance (UXO) cleanup is the number one priority Army Cleanup Problem requirement and is identified as a major problem throughout the Department of Defense (DOD). A recent SERDP technical report summarizes the status of capability for buried UXO detection, discrimination, and identification as follows: (a) can detect UXO, within definable limits; (b) cannot effectively discriminate UXO anomalies from ‘false alarm’ anomalies; and (c) cannot identify UXO. False alarm anomalies are defined here as geophysical anomalies caused by buried UXO debris, other metallic objects, gravel and cobbles, soil heterogeneities, tree roots, and other natural and cultural features. False alarm anomalies significantly contribute to the cost of UXO remediation due to the large number of unnecessary excavations, A major initiative in the research and development community, therefore, is to develop discrimination (target identification) capabilities. One potential methodology for target identification involves utilizing the broadband scattered electromagnetic induction response. This technique, which is known as Electromagnetic Induction Spectroscopy (EMIS), h as recently become feasible due to the development of the GEM-3 sensor. The GEM-3 is an efficient, broadband, handheld EMI sensor than employs a unique monostatic coil design. Analyzing the EMI spectral content for target identification is not new. In fact, elementary EM theory states that an object must exhibit different responses at different frequencies. All fundamental EM equations involving a timevarying source testify as such. By fully characterizing and identifying an object without excavation, we should be able to significantly reduce the number of false targets. EMIS should be fully applicable to many other problems where target identification and recognition (without intrusive search) are important.