Tesselations by Connection in Orders

  • Authors:
  • Michel Couprie;Gilles Bertrand

  • Affiliations:
  • -;-

  • Venue:
  • DGCI '00 Proceedings of the 9th International Conference on Discrete Geometry for Computer Imagery
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

The watershed transformation is a powerful tool for segmenting images, but its precise definition in discrete spaces raises difficult problems. We propose a new approach in the framework of orders. We introduce the tesselation by connection, which is a transformation that preserves the connectivity, andcan be implementedb y a parallel algorithm. We prove that this transformation possesses goodg eometrical properties. The extension of this transformation to weighted orders may be seen as a generalization of the watershed transformation.