Share Email Print

Proceedings Paper

If training data appears to be mislabeled, should we relabel it? Improving supervised learning algorithms for threat detection in ground penetrating radar data
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

This work focuses on the development of automatic buried threat detection (BTD) algorithms using ground penetrating radar (GPR) data. Buried threats tend to exhibit unique characteristics in GPR imagery, such as high energy hyperbolic shapes, which can be leveraged for detection. Many recent BTD algorithms are supervised, and therefore they require training with exemplars of GPR data collected over non-threat locations and threat locations, respectively. Frequently, data from non-threat GPR examples will exhibit high energy hyperbolic patterns, similar to those observed from a buried threat. Is it still useful therefore, to include such examples during algorithm training, and encourage an algorithm to label such data as a non-threat? Similarly, some true buried threat examples exhibit very little distinctive threat-like patterns. We investigate whether it is beneficial to treat such GPR data examples as mislabeled, and either (i) relabel them, or (ii) remove them from training. We study this problem using two algorithms to automatically identify mislabeled examples, if they are present, and examine the impact of removing or relabeling them for training. We conduct these experiments on a large collection of GPR data with several state-of-the-art GPR-based BTD algorithms.

Paper Details

Date Published: 30 April 2018
PDF: 9 pages
Proc. SPIE 10628, Detection and Sensing of Mines, Explosive Objects, and Obscured Targets XXIII, 106280F (30 April 2018); doi: 10.1117/12.2305881
Show Author Affiliations
Daniël Reichman, Duke Univ. (United States)
Leslie M. Collins, Duke Univ. (United States)
Jordan M. Malof, Duke Univ. (United States)

Published in SPIE Proceedings Vol. 10628:
Detection and Sensing of Mines, Explosive Objects, and Obscured Targets XXIII
Steven S. Bishop; Jason C. Isaacs, Editor(s)

© SPIE. Terms of Use
Back to Top