An Efficient Local and Global Model for Image Segmentation

Quang Tung Thieu, Marie Luong, Jean-Marie Rocchisani, Dat Tran, Emmanuel Viennet

Research output: A Conference proceeding or a Chapter in BookConference contributionpeer-review

2 Citations (Scopus)


In this paper, a new region-based active contour model using a variational level set formulation is proposed for image segmentation. The model is based on curve evolution, local statistical function and level set method. The energy function for the proposed model consists of two components: global component and local component. By introducing the local term, the images with intensity inhomogeneities can be efficiently segmented. Moreover, a smoothness regularization is derived from a Gaussian filtering term. This allows avoiding re-initialization while ensuring the smoothness of the level set function. The addition of the global term makes the model more flexible to the location of initial contour. Experimental results show that our method is less sensitive to the location of initial contour and demonstrate the performance of our model
Original languageEnglish
Title of host publicationThe 2011 International Conference on Advanced Technologies for Communications
EditorsDuy-Hieu Bui, Xuan-Tu Tran
Place of PublicationPiscataway USA
Number of pages4
ISBN (Electronic)9781457712067
ISBN (Print)9781457712074
Publication statusPublished - 2011
EventInternational Conference on Advanced Technologies for Communications (ATC): (ATC 2011) - Da Nang University of Technology, Hanoi, Viet Nam
Duration: 2 Aug 20114 Aug 2011 (Conference detail)


ConferenceInternational Conference on Advanced Technologies for Communications (ATC)
Abbreviated titleATC
Country/TerritoryViet Nam
OtherATC fosters an international forum for scientific and technological exchange among scientists and engineers in the fields of electronics, communications and related areas. It gives a focus on advanced techniques for high-speed communications
Internet address


Dive into the research topics of 'An Efficient Local and Global Model for Image Segmentation'. Together they form a unique fingerprint.

Cite this