DocumentCode :
2775462
Title :
Segmentation of deformable templates with level sets characterized by particle systems
Author :
Yahia, H.M. ; Berroir, J.P.
Author_Institution :
Inst. Nat. de Recherche en Inf. et Autom., Le Chesnay, France
Volume :
2
fYear :
1998
fDate :
16-20 Aug 1998
Firstpage :
1421
Abstract :
A method for the segmentation of deformable structures using a formulation of level-sets with particles system is presented. This results in an original energetic formulation, and leads to a fast and robust method for computing the segmentation. Initialization is performed with the help of skeleton techniques, providing a stable and coherent initialization. It is possible to add or delete particles during the minimization process, and the use of particles systems makes it possible to contemplate hierarchical representations. The global shape is then represented by few control points, and more precision is obtained by locally adding new particles
Keywords :
feature extraction; image representation; image segmentation; minimisation; polynomials; contour extraction; deformable templates; image segmentation; initialization; level set method; minimization; particle systems; polynomials; shape representation; skeleton techniques; Automatic control; Computer graphics; Control systems; Electrical capacitance tomography; Level set; Minimization methods; Polynomials; Potential energy; Shape; Spline;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on
Conference_Location :
Brisbane, Qld.
ISSN :
1051-4651
Print_ISBN :
0-8186-8512-3
Type :
conf
DOI :
10.1109/ICPR.1998.711969
Filename :
711969
Link To Document :
بازگشت