Skip to main navigation Skip to search Skip to main content

Interactive tongue body segmentation

  • School of Computer Science and Technology, Harbin Institute of Technology

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

Abstract

Despite the progress in tongue body segmentation, it remains a challenge to correctly segment the tongue body from any tongue images. As one possible solution, conventional interactive segmentation methods usually are not tailored for tongue body segmentation, and suffer from several limitations like poor efficiency and less specificity. In this paper, we extend the GaborFM method, and propose an interactive tongue body segmentation (ITBS) method by considering three principles: convergence, efficiency, and ease of use. Using the segmentation result of GaborFM for initialization, the proposed method allows the user to modify the segmentation by simply drawing a line segment at the true tongue body contour. Then ITBS exploits the information provided by the user to update the masks and edges, and update contour using fast marching. Experimental results show that ITBS can obtain correct results after less than 0.5 times of interactions in average, and is much efficient than conventional interactive segmentation methods.

Original languageEnglish
Title of host publicationProceedings - 2014 International Conference on Medical Biometrics, ICMB 2014
PublisherIEEE Computer Society
Pages26-31
Number of pages6
ISBN (Print)9781479940141
DOIs
StatePublished - 2014
Externally publishedYes
Event2014 International Conference on Medical Biometrics, ICMB 2014 - Shenzhen, Guangdong, China
Duration: 30 May 20141 Jun 2014

Publication series

NameProceedings - 2014 International Conference on Medical Biometrics, ICMB 2014

Conference

Conference2014 International Conference on Medical Biometrics, ICMB 2014
Country/TerritoryChina
CityShenzhen, Guangdong
Period30/05/141/06/14

Keywords

  • GaborFM
  • Tongue diagnosis
  • interactive tongue body segmentation

Fingerprint

Dive into the research topics of 'Interactive tongue body segmentation'. Together they form a unique fingerprint.

Cite this