New Models and Fast Algorithms for Natural and Urban Clutter with Applications

Abstract

To address the Automatic Target Detection/Recognition (ATD/R) community's long term goal of 'revolutionizing wide-area imagery analysis', the research presented in this report focused on two enabling technologies: (1) clutter models based on Synthetic Aperture Radar (SAR) in urban environments; and (2) fast algorithms for Markov Random Fields. The former investigated the nature of building signatures in SAR imagery, and saw the development of a building detector. Buildings contribute a large number of false target detections, and targets stationed in close proximity to buildings can be missed using conventional analysis methods. The latter effort revisited signal processing to make theoretical headway in developing reduced computational cost techniques for estimating and using stochastic target and clutter models based on Markov Random Fields. These algorithms have broad application to a large variety of ATD/R concerns.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1997
Accession Number
ADA332033

Entities

People

  • David B. Cooper

Organizations

  • Brown University

Tags

Communities of Interest

  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Detection
  • Detectors
  • Equations
  • Estimators
  • False Alarms
  • Frequency Domain
  • Geometry
  • Image Processing
  • Probability
  • Random Variables
  • Signal Processing
  • Stochastic Processes
  • Synthetic Aperture Radar
  • Target Detection
  • Two Dimensional
  • Warning Systems

Fields of Study

  • Engineering

Readers

  • Computational Modeling and Simulation
  • Radar Systems Engineering.
  • Sensor Fusion and Tracking Systems.