Automatic Target Recognition, Executive Summary and Annotated Brief (PR)

Abstract

The complexity of warfare and the requirement to shorten the kill chain against difficult targets has increased the need for Automatic Target Recognition (ATR). No single sensor currently provides a robust enough capability of detecting all classes of targets through the camouflage and deceptive elements of natural and/or man-made clutter. ATR is a sub-element of a larger class of problems known collectively as Target Identification. This study concentrated on image-based techniques as an ATR subdiscipline of target ID. The study panel concluded that no silver bullet exists that will dramatically improve ATR performance in the near term. However, more work and understanding in very specific areas could improve the way the Air Force uses ATR today. For example, CONOPS and technology can work synergistically (e.g. change detection) to improve mission capabilities; ATR can be an aid to, but not a substitute for, human judgment by both pilots and image analysts; and ATR is essential to better match image exploitation capabilities with collection capacity.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 2005
Accession Number
ADA633787

Entities

People

  • Llewellyn Dougherty

Tags

Communities of Interest

  • Autonomy
  • Human Systems
  • Sensors
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Automated Target Recognition
  • Change Detection
  • Cognition
  • Databases
  • Detection
  • Detectors
  • Guidance
  • Inertial Navigation
  • Inertial Navigation Systems
  • Military Research
  • Military Science
  • Navigation
  • Target Recognition
  • Test And Evaluation
  • Three Dimensional
  • Warfare

Readers

  • Sensor Fusion and Tracking Systems.
  • Systems Analysis and Design