Va-3092-003-02-tr Serdp Project Cu-1121 Final Report Processing Techniqes for Discrimination between Buried Uxo and Clutter Using Multisensor Array Data

Although commercially available geophysical sensors are capable of detecting UXO at nominalburial depths, they cannot reliably discriminate between UXO and clutter. As a result, anestimated 75% of remediation funds are spent on nonproductive excavations. During the pastfew years, we have been studying the merits of using multifrequency EMI data fordiscriminating between UXO and non-UXO targets and believe the method has tremendouspotential. The EMI spectral response of an object is a function of its electrical conductivity,magnetic permeability, shape, size, and orientation relative the primary exciting field. Bymeasuring a target’s spectral response, we obtain its characteristic frequency-dependentsignature. Based on the response spectrum, we “fingerprint” the object and compare its responseto known UXO signatures.To explore this phenomenon, we have developed a unique, frequency-domain EMI sensor namedthe GEM-3, which operates over a bandwidth of 30 Hz to 24 kHz. Empirical data acquired usingthe GEM-3 for a wide assortment of UXO and non-UXO suggests strongly that the EMIanomaly measured in a broad band offers an ability to both detect and identify a target. Wepresent results of controlled measurements made with the GEM-3 sensor to address the questionof competing effects such as sensor stability, depth and shape effects, and inhomogeneousmaterial properties. The frequency-dependent signatures are reproducible and consistent over arange of depths. Therefore, we have established that the GEM-3 is capable of deliveringbroadband EMI data with ample target-specific information for the purpose of targetclassification and identification.