Comparison of Spatial Frequency Domain Features for the Detection of Side Attack Explosive Ballistics in Synthetic Aperture Acoustics

Dowdy, J., Anderson, D., Luke, R., Ball, J. E., Havens, T., & Keller, J. M. (2016). Comparison of Spatial Frequency Domain Features for the Detection of Side Attack Explosive Ballistics in Synthetic Aperture Acoustics. SPIE Defense and Commercial Sensing Conference 2016. Baltimore, MD: SPIE.

Abstract

Explosive hazards in current and former conflict zones are a serious threat to both civilians and soldiers alike. Herein, we explore the detection of side attack explosive ballistics via a vehicle-mounted acoustic sensor. In particular, we explore different spatial frequency domain transformations and features on digitally beamformed acoustic data and also on the raw acoustic returns. These features are explored using GAMKLp-SVM, a genetic algorithm-based multiple kernel learning technique and SVM classifier. Performance is assessed using ROC curves on data from a U.S. Army test site that contains multiple target and clutter types, burial depths and times of day.