Texture Classification Using Averages of Local Pattern Matches

Abstract

Laws has introduced a class of texture features based on average degrees of match of the pixel neighborhoods with a set of standard masks. These features yield better texture classification than standard features based on pairs of pixels. This paper investigates simpifications of these features, and shows that their performance is not greatly affected by their exact form, and also appears to remain the same if only local match maxima are used. It also presents an alternative definition of such features based on sums and differences of Gaussian convolutions.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1981
Accession Number
ADA124807

Entities

People

  • Azriel Rosenfeld
  • Larry S. Davis
  • Matti Pietikaeinen

Organizations

  • University of Maryland

Tags

Communities of Interest

  • Energy and Power Technologies
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Change Detection
  • Classification
  • Computer Science
  • Computer Vision
  • Computers
  • Contracts
  • Convolution
  • Detection
  • Detectors
  • Electrical Engineering
  • Image Processing
  • Night Vision
  • Order Statistics
  • Pattern Recognition
  • Standards
  • Statistics
  • Universities

Readers

  • Approximation Theory.
  • Radar Systems Engineering.
  • Theoretical Analysis.