A Mathematical Framework for Image Analysis

Abstract

The results reported here were derived from the research project 'A Mathematical Framework for Image Analysis' supported by the Office of Naval Research, contract N00014-88-K-0289 to Brown University. A common theme for the work reported is the use of probabilistic methods for problems in image analysis and image reconstruction. Five areas of research are described: rigid body recognition using a decision tree/combinatorial approach; nonrigid body recognition using deformable templates; the construction of two and three dimensional shape models for complex recognition and interpretation tasks; three dimensional shape reconstruction; and texture analysis.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1991
Accession Number
ADA258277

Entities

People

  • Basilis Gidas
  • Donald E. Mcclure
  • Donald Geman
  • Stuart Geman
  • Ulf Grenander

Organizations

  • Brown University

Tags

Communities of Interest

  • Advanced Electronics
  • Biomedical
  • Engineered Resilient Systems
  • Ground and Sea Platforms
  • Human Systems
  • Sensors
  • Space

DTIC Thesaurus Topics

  • Algorithms
  • Character Recognition
  • Computational Science
  • Computer Programming
  • Computer Vision
  • Computers
  • Differential Equations
  • Hidden Markov Models
  • Image Processing
  • Markov Models
  • Mathematical Models
  • Object Recognition
  • Probabilistic Models
  • Random Variables
  • Stationary Processes
  • Synthetic Aperture Radar
  • Three Dimensional

Readers

  • Computer Vision.
  • Technical Research and Report Writing.