Applications of Nonlinear Adaptive Filters for Image Enhancement,

Abstract

Generalizations of median filters which combine desirable properties of both linear and nonlinear filters have recently been developed by the authors. This paper gives some results of applications of these filters in the enhancement of noisy images. The authors consider in particular the median, the alpha-trimmed mean (alpha-TM), the modified trimmed mean (MTM) and the double window (DW) MTM filters. In all but the last case iterated use of the filter has been examined. The results show that the new filters (MTM and DW MTM filters) are very good for edge-preserving enhancement of images contaminated by additive noise which includes impulsive components.

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1984
Accession Number
ADA151322

Entities

People

  • S. A. Kassam
  • Y. H. Lee

Organizations

  • Moore School of Electrical Engineering

Tags

DTIC Thesaurus Topics

  • Adaptive Filters
  • Additives (Chemicals)
  • Filters
  • Pattern Recognition
  • Recognition

Readers

  • Computer Vision.
  • Image Processing and Computer Vision.
  • Statistical inference.