A New Interpretation and improvement of the Nonlinear Anisotropic Diffusion for Image Enhancement

  • Authors:
  • Jérôme Monteil;Azeddine Beghdadi

  • Affiliations:
  • Univ. Paris Nord, Villetaneuse, France;Univ. Paris Nord, Villetaneuse, France

  • Venue:
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Year:
  • 1999

Quantified Score

Hi-index 0.15

Visualization

Abstract

The purpose of this article is to give a new analysis of the Anisotropic Diffusion (AD) and propose an adaptive nonlinear filtering based on a judicious choice of the Conductance Function (CF) and the edgeness threshold. A new undesirable effect, which we call the 驴pinhole effect,驴 may result when AD is introduced for the first time. A robust solution to this effect is proposed and evaluated through experimental data. The evolution of the diffused signal is analyzed through a physical model using the Optical Flow Technique (OFT). The overall strategy is evaluated through experimental results obtained on synthetic and actual images.