A stochastic evaluation of the contour strength

Abstract : If one considers only local neighborhoods for segmenting an image, one gets contours whose strength is often poorly estimated. A method for reevaluating the contour strength by taking into account non local features is presented: one generates a fixed number of random germs which serve as markers for the watershed segmentation. For each new population of markers, another set of contours is generated. "Important" contours are selected more often. The present paper shows that the probability that a contour is selected can be estimated without performing the effective simulations. Copyright Springer-Verlag 2010. The original publication is available at www.springerlink.com/content/y057x103475301r2/
Type de document :
Communication dans un congrès
Goesele, M.; Roth, S.; Kuijper, A.; Schiele, B.; Schindler, K. 32nd Annual Symposium of the German Association for Pattern Recognition (DAGM 2010), Sep 2010, Darmstadt, Germany. Springer Berlin / Heidelberg, 6376/2010, pp.513-522, 2010, Lecture Notes in Computer Science: Image Processing, Computer Vision, Pattern Recognition, and Graphics. 〈10.1007/978-3-642-15986-2_52〉
Liste complète des métadonnées

Littérature citée [7 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-00614592
Contributeur : Cybèle Ciofolo-Veit <>
Soumis le : samedi 12 novembre 2011 - 07:00:05
Dernière modification le : vendredi 27 octobre 2017 - 17:36:02
Document(s) archivé(s) le : lundi 13 février 2012 - 02:20:07

Fichier

Stawiaski_DAGM2010.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Fernand Meyer, Jean Stawiaski. A stochastic evaluation of the contour strength. Goesele, M.; Roth, S.; Kuijper, A.; Schiele, B.; Schindler, K. 32nd Annual Symposium of the German Association for Pattern Recognition (DAGM 2010), Sep 2010, Darmstadt, Germany. Springer Berlin / Heidelberg, 6376/2010, pp.513-522, 2010, Lecture Notes in Computer Science: Image Processing, Computer Vision, Pattern Recognition, and Graphics. 〈10.1007/978-3-642-15986-2_52〉. 〈hal-00614592〉

Partager

Métriques

Consultations de la notice

167

Téléchargements de fichiers

92