Performance Modelling of Autonomous Electro-Optical Sensors. Experimental Algorithms and Software for ATR

Abstract

We shall study the problem of object identification against a background of clutter and using laser radar for obtaining range data. The algorithm shall attempt to classify the object or objects in the scene. It may also be able to do the detection but this is not the primary goal since, in a multiple sensor situation, this may be better done by other sensors. The algorithms shall be flexible so that they can easily be modified to incorporate changes in technology (as long as range data are supplied). In particular, if the accuracy of the range determinations improves drastically, only minor changes will be needed in the code. To make this possible the code is made highly modular, so that only one, or a few, of the smaller modules need be replaced by other modules (also supplied) in addition, of course, to changing parameter values.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1987
Accession Number
ADA230948

Entities

People

  • Ulf Grenander

Tags

Communities of Interest

  • Air Platforms
  • C4I
  • Energy and Power Technologies
  • Human Systems
  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Bayesian Networks
  • Computational Science
  • Data Analysis
  • Detection
  • Detectors
  • Identification
  • Information Science
  • Laser Radar
  • Mathematical Models
  • Probability
  • Probability Distributions
  • Recognition
  • Simulators
  • Statistical Algorithms
  • Statistical Analysis
  • Statistics

Readers

  • Image Processing and Computer Vision.
  • Software Engineering
  • Systems Analysis and Design

Technology Areas

  • Directed Energy