Subsurface target imaging using a multi-resolution 3D quadtree algorithm

Abstract

ABSTRACT The imaging of subsurface targets, such as landmines, using Ground Penetrating Radar (GPR) is becomingan increasingly important area of research. Conventional image formation techniques expend large amounts ofcomputational resources on fully resolving a region, even if there is a large amount of clutter. For example,standard backprojection algorithms require O (N 3 ). However, by using multi-resolution techniques - such asquadtree - potential targets and clutter can be discriminated more eciently with O (N 2 log 2 N ). Because priorwork has focused on the imaging of surface targets, quadtree techniques have mostly been developed for 2Dimaging. Target depth adds another dimension to the imaging problem; therefore, we have developed a 3Dquadtree algorithm. In this case, the mine “eld is modeled as a volume that is sub-divided at each stage ofthe quadtree algorithm. From each of these sub-volumes, the energy intensity is calculated. As the algorithmproceeds to “ner resolutions, the energy in region containing a potential target increases, while that of backgroundnoise decreases. A multi-stage detector applied on intermediate quadtree data uses this change in energy todiscriminate between regions of targets and clutter. This is advantageous because only the regions containinglikely targets are investigated by additional sensors that are relatively slow in comparison to GPR (e.g. seismicor EMI sensors). This algorithm is tested on synthetic and experimental data collected from a model mine “eldat Georgia Institute of Technology. Even under near “eld and small aperture conditions, which hold for the minedetection case, test results show that target location information can be gathered with processing using the 3Dquadtree algorithm.Keywords: Quadtree Backprojection, Subsurface Imaging, 3D Imaging, Mine Detection, Ground PenetratingRadar