A Model of an Expert Computer Vision and Recognition Facility with Applications of a Proportion Technique.

Abstract

In this thesis an expert vision system is proposed which infers, from the collection and evaluation of facial segments of a face, a set of possible identifies. Retrievals from a data store to accomplish this are facilitated by using the shape representations as keys to the records. The heuristic involves collecting the candidate identities in those sets by set intersection and a weighting scheme. Sensitivity and tolerance factors in this expert are adjustable in terms of range values. The value obtained during evaluation of a facial segment is accepted or tolerated relative to the range of values expected for that segment. A retrieval effects a subrange whose width is determined by sensitivity to be the center and some number of adjacent values above and below that center value. Expectation ranges can be tailored to incorporate system and environmental variables.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 15, 1985
Accession Number
ADA158073

Entities

People

  • G. E. Sherman

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Computational Science
  • Computer Science
  • Computer Vision
  • Computers
  • Databases
  • Feature Extraction
  • Identification
  • Image Processing
  • Image Recognition
  • Information Science
  • Information Systems
  • Pattern Recognition
  • Processing Equipment
  • Recognition
  • Self Organizing Systems
  • United States Military Academy

Readers

  • Artificial Intelligence
  • Control Systems Engineering.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference