Summary of the DARPA background clutter data collection experiment

Most technologies in use or proposed for use to detect landmines and unexploded ordnance (UXO) suffer from unacceptably high false-alarm rates, even at modest probabilities of detection. High false-alarm rates are a consequence of the inability to discriminate real UXO and landmines from man-made and naturally occurring clutter. Experimental data on expected signatures of UXO and landmines have been collected by various sensor developers and research agencies. However, an equivalent characterization of clutter returns has been lacking. As a result, background signatures have not been considered in sensor or algorithm design. The goal of the two-phase, DARPA-sponsored background clutter data collection experiment is to begin to fill this void. The first phase, completed during the fall of 1996, consisted of high aerial density site surveys using the following sensor types: magnetometer, infrared, electromagnetic induction, and ground-penetrating radar. The second phase, which is currently underway, consists of analysis of the collected sensor data to include excavation of clutter objects. This paper describes the clutter data collection program and provides a cursory summary of the analysis done to date.