Histogram based segmentation using Wasserstein distances

Tony Chan*, Selim Esedoglu, Kangyu Ni

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

35 Scopus citations


In this paper, we propose a new nonparametric region-based active contour model for clutter image segmentation. To quantify the similarity between two clutter regions, we propose to compare their respective histograms using the Wasserstein distance. Our first segmentation model is based on minimizing the Wasserstein distance between the object (resp. background) histogram and the object (resp. background) reference histogram, together with a geometric regularization term that penalizes complicated region boundaries. The minimization is achieved by computing the gradient of the level set formulation for the energy. Our second model does not require reference histograms and assumes that the image can be partitioned into two regions in each of which the local histograms are similar everywhere.

Original languageEnglish (US)
Title of host publicationScale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings
PublisherSpringer Verlag
Number of pages12
ISBN (Print)9783540728221
StatePublished - 2007
Externally publishedYes
Event1st International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2007 - Ischia, Italy
Duration: May 30 2007Jun 2 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4485 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other1st International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2007


  • Clutter
  • Image segmentation
  • Region-based active contour
  • Wasserstein distance

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Histogram based segmentation using Wasserstein distances'. Together they form a unique fingerprint.

Cite this