Efficient multi-object segmentation of 3D medical images using clustering and graph cuts

Abstract : We propose an application of multi-label ''Graph Cut" optimization algorithms to the simultaneous segmentation of multiple anatomical structures, initialized via an over-segmentation of the image computed by a fast centroidal Voronoi diagram (CVD) clustering algorithm. With respect to comparable segmentations computed directly on the voxels of image volumes, we demonstrate performance improvements on both execution speed and memory footprint by, at least, an order of magnitude, making it possible to process large volumes on commodity hardware which could not be processed pixel-wise.
Type de document :
Communication dans un congrès
18th IEEE International Conference on Image Processing (ICIP 2011), Sep 2011, Bruxelles, Belgium. pp.2149-2152, 2011, <10.1109/ICIP.2011.6116036>


https://hal.archives-ouvertes.fr/hal-00658028
Contributeur : Sébastien Valette <>
Soumis le : lundi 9 janvier 2012 - 17:04:58
Dernière modification le : mardi 17 novembre 2015 - 17:31:26
Document(s) archivé(s) le : mardi 10 avril 2012 - 02:37:35

Fichiers

kechichian-et-al-icip-2011.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Razmig Kéchichian, Sébastien Valette, Michel Desvignes, Rémy Prost. Efficient multi-object segmentation of 3D medical images using clustering and graph cuts. 18th IEEE International Conference on Image Processing (ICIP 2011), Sep 2011, Bruxelles, Belgium. pp.2149-2152, 2011, <10.1109/ICIP.2011.6116036>. <hal-00658028>

Exporter

Partager

Métriques

Consultations de
la notice

267

Téléchargements du document

207