APPLICATION OF RECOGNITION THEORY TO MISSILE IDENTIFICATION AND DECOY DISCRIMINATION

Abstract

The problem of categorizing numerically described events is discussed in terms of a set of categories defined by correctly identified sample events. From the point of view of statistical decision-theoretical methods, classification and learning through parameter estimation constitute what may be called pattern recognition. The statistical classification techniques discussed can be used (and are optimum, for a particular criterion) when there is adequate prior knowledge about the functional forms of the category distributions. Nonparametric techniques are used when there is little prior knowledge. Some of these techniques have already been implemented and tried on real speech data. The results of this limited trial were successful. (Author)

Document Details

Document Type
Technical Report
Publication Date
May 12, 1961
Accession Number
AD0284854

Entities

Organizations

  • Melpar

Tags

Communities of Interest

  • Weapons Technologies

DTIC Thesaurus Topics

  • Classification
  • Discrimination
  • Identification
  • Learning
  • Pattern Recognition
  • Recognition

Readers

  • Artificial Intelligence
  • Neural Network Machine Learning.
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks