Performance of an adaptive feature-based processor for a wideband ground penetrating radar system


Journal Article

A two-stage algorithm for landmine detection with a ground penetrating radar (GPR) system is described. First, 3-D data sets are processed using a computationally inexpensive pre-screening algorithm which flags potential locations of interest. These flagged locations are then passed to a feature-based processer which further discriminates target-like anomalies from naturally occurring clutter. Current field trial (over 6500 square meters) and blind test results (over 39000 square meters) are presented and these show at least an order of magnitude improvement over other radar system-based detection algorithms on the same test lanes. © 2006 IEEE.

Full Text

Duke Authors

Cited Authors

  • Torrione, PA; Throckmorton, CS; Collins, LM

Published Date

  • April 1, 2006

Published In

Volume / Issue

  • 42 / 2

Start / End Page

  • 644 - 658

International Standard Serial Number (ISSN)

  • 0018-9251

Digital Object Identifier (DOI)

  • 10.1109/TAES.2006.1642579

Citation Source

  • Scopus