Multinomial Level-Set Framework for Multi-region Image Segmentation

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

Abstract

We present a simple and elegant level-set framework for multi-region image segmentation. The key idea is based on replacing the traditional regularized Heaviside function with the multinomial logistic regression function, commonly known as Softmax. Segmentation is addressed by solving an optimization problem which considers the image intensities likelihood, a regularizer, based on boundary smoothness, and a pairwise region interactive term, which is naturally derived from the proposed formulation. We demonstrate our method on challenging multi-modal segmentation of MRI scans (4D) of brain tumor patients. Promising results are obtained for image partition into the different healthy brain tissues and the malignant regions.
Original languageEnglish
Title of host publicationScale Space and Variational Methods in Computer Vision - 6th International Conference, SSVM 2017, Proceedings
EditorsFrancois Lauze, Yiqiu Dong, Anders Bjorholm Dahl
PublisherSpringer Verlag
Pages386-395
Number of pages10
ISBN (Print)9783319587707
DOIs
StatePublished - 2017
Event6th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2017 - Kolding, Denmark
Duration: 4 Jun 20178 Jun 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10302 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference6th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2017
Country/TerritoryDenmark
CityKolding
Period4/06/178/06/17

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Multinomial Level-Set Framework for Multi-region Image Segmentation'. Together they form a unique fingerprint.

Cite this