A Priority-Driven General Approach of Segmentation-Recognition of Connected Symbols

Christophe Renaudin 1 Yann Ricquebourg 1 Jean Camillerapp 1
1 IMADOC - Interprétation et Reconnaissance d’Images et de Documents
UR1 - Université de Rennes 1, INSA Rennes - Institut National des Sciences Appliquées - Rennes, CNRS - Centre National de la Recherche Scientifique : UMR6074
Abstract : In this paper, we present a general method to segment and recognize two or more connected symbols. The method is based on the evaluation of segmentation candidates produced by grouping elements of an over-segmentation. The possible segmentation areas obtained by the over-segmentation are hierarchically ordered, which allows a priority-driven approach in the investigation of correct segmentation. The method was tested on connected handwritten digits and on connected printed mathematical symbols.
Type de document :
Communication dans un congrès
11th International Conference on Frontiers in Handwriting Recognition - ICFHR 2008, Aug 2008, Montréal, Canada. p. 629-634, 2008
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00326050
Contributeur : Christophe Renaudin <>
Soumis le : mercredi 1 octobre 2008 - 13:03:26
Dernière modification le : vendredi 16 novembre 2018 - 01:23:45

Identifiants

  • HAL Id : hal-00326050, version 1

Citation

Christophe Renaudin, Yann Ricquebourg, Jean Camillerapp. A Priority-Driven General Approach of Segmentation-Recognition of Connected Symbols. 11th International Conference on Frontiers in Handwriting Recognition - ICFHR 2008, Aug 2008, Montréal, Canada. p. 629-634, 2008. 〈hal-00326050〉

Partager

Métriques

Consultations de la notice

114