Multi-Frequency Analysis for Landmine Detection with Forward-Looking Ground Penetrating Radar

Abstract

The project investigates image processing, sensor fusion and signal processing techniques for the forward-looking ground penetrating radar (FLGPR) explosive detection system equipped with a color or FLIR camera (the Alaric system fielded by NVESD), as well as independent multi-camera systems. Also, in this report period, we are addressing research issues dealing with feature and sensor fusion. We had some partial funding from a Leonard wood Institute Grant. The ultimate goal is to utilize multiple sensing modalities together with FLGPR to increase IED detection with low false alarm rates. The project objectives are to: Perform image processing for infra-red and color cameras to detect surface laid road-side targets; Investigate advanced target detection approaches for the FLGPR; Develop coordinate mapping technique between EO image sensors and FLGPR data and investigate fusion algorithms; Research and develop approaches for vehicle-based human-in-the-loop cuing of explosive devices using EO sensors; and Examine and process the EO and FLGPR data collected by the U.S. Army and improve algorithm performance through extensive testing.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 12, 2010
Accession Number
ADA532712

Entities

People

  • Dominic K. Ho
  • James M. Keller

Tags

Communities of Interest

  • Counter IED
  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Anomaly Detection
  • Change Detection
  • Computer Vision
  • Data Sets
  • Detection
  • Detectors
  • Explosive Devices
  • False Alarms
  • Feature Extraction
  • Image Processing
  • Machine Learning
  • Pattern Recognition
  • Sensor Fusion
  • Students
  • Supervised Machine Learning
  • Warning Systems

Readers

  • Computer Vision.
  • Sensor Fusion and Tracking Systems.