Amoeba active contours

  • Authors:
  • Martin Welk

  • Affiliations:
  • Institute for Biomedical Image Analysis, University for Health Sciences, Medical Informatics and Technology, Hall/Tyrol, Austria

  • Venue:
  • SSVM'11 Proceedings of the Third international conference on Scale Space and Variational Methods in Computer Vision
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

We introduce an algorithm for active contour segmentation in which the level set function encoding the contour is processed by median filtering using morphological amoebas. These are adaptive structure elements introduced by Lerallut et al. which can be combined with different morphological operations. Recently it has been proven that iterated amoeba median filtering of an image approximates the well-known self-snakes partial differential equation. Following this approach we prove a partial approximation property of amoeba active contours with respect to geodesic active contours. Experiments prove the viability of the algorithm and confirm the theoretical results.