AUTOMATIC TARGET RECOGNITION DEVICE (ATRD).

Abstract

Military targets imaged on reconnaissance aerial photography were automatically identified by parametric pattern recognition techniques implemented in an Automatic Target Recognition Device. Successful performance of this equipment over a 16-class strategic and tactical target recognition requirement has significant implications to the field of military photo interpretation. Target characteristics are represented in statistical form by operating on the video waveform representation of a target area. Clue extraction, machine learning, and target recognition are accomplished under computer control; target decisions and reliabilities appear as a typed output. A comprehensive test and evaluation phase involving 350 frames of aerial photography yield classification error rates from 0.05 to 0.25. (Author)

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1966
Accession Number
AD0478454

Entities

People

  • G. O. Richards
  • J. W. Gerdes
  • W. B. Floyd

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Aerial Photography
  • Classification
  • Machine Learning
  • Pattern Recognition
  • Photographic Equipment
  • Photographic Materials
  • Photography
  • Recognition
  • Reconnaissance
  • Target Recognition
  • Test And Evaluation

Fields of Study

  • Engineering

Readers

  • Computer Science/Computer Engineering/Data Science/Digital Signal Processing.
  • Computer Vision.

Technology Areas

  • AI & ML