Theoretical Analysis of Models for Texture

Abstract

During the period covered by this contract, technical advances were made in the general area of describing image textures in terms of the spatial distribution of local features, such as edges, in the texture. The first contribution concerned the theoretical development of a minimal error one dimensional edge detector for image models which can be used to describe textures. The second contribution was a comparative classification study of texture statistics derived form grey level and edge cooccurrence matrices. The third contribution was the development of a new computational tool for analyzing textures, called a polarogram. The polarogram is a rich source of directionally sensitive texture statistics which are invariant to image orientation. This research was documented in seven reports and papers.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1980
Accession Number
ADA208486

Entities

People

  • Larry S. Davis

Organizations

  • University of Texas at Austin

Tags

Communities of Interest

  • Autonomy
  • Human Systems
  • Weapons Technologies

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Change Detection
  • Classification
  • Computer Graphics
  • Computer Science
  • Computers
  • Decision Theory
  • Detection
  • Detectors
  • Graphics
  • Image Processing
  • Mathematics
  • Orientation (Direction)
  • Pattern Recognition
  • Spatial Distribution
  • Statistics

Readers

  • Computer Vision.
  • Theoretical Analysis.