Investigation of Coherent and Incoherent Change Detection Algorithms

Abstract

An investigation of five change detection methods used in synthetic aperture radar (SAR) is presented in this thesis. This investigation utilizes data gathered from the Air Force Research Laboratory (AFRL) Sensor Data Management System (SDMS) in order to compare the various change detection techniques. These change detection methods include the following: a) incoherent change detection (ICCD), b) coherent change detection (CCD), c) alternative coherent change detection (ACCD), d) log likelihood change statistic (LLCS), and e) a two-stage change detection, which involves a combination of ICCD and CCD. In addition, a new change detection method for comparison with these five basic methods is developed. This investigation reveals that the LLCS statistic is the most promising method for revealing changes within the SDMS dataset. Furthermore, the authors change detection method yields overall visual improvement in comparison to the two-stage change detection method.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 2017
Accession Number
AD1053496

Entities

People

  • Nicholas S. Underwood

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Energy and Power Technologies
  • Ground and Sea Platforms
  • Sensors

DTIC Thesaurus Topics

  • Air Force
  • Air Force Research Laboratories
  • Algorithms
  • Change Detection
  • Data Management
  • Detection
  • Detectors
  • Electrical Engineering
  • False Alarms
  • Fungi
  • Military Research
  • Probability
  • Radar
  • Random Variables
  • Simulators
  • Synthetic Aperture Radar
  • Warning Systems

Readers

  • Computer Vision.
  • Educational Psychology
  • Regression Analysis.