An Error-Correction Graph Grammar to Recognize Textured Symbols

Abstract : This paper presents an algorithm for recognizing symbols with textured elements in a graphical document. A region adjacency graph represents the graphical document, with the nodes being polygons and the edges the neighborhood relations between them. The textured symbols are modeled by a graph, where nodes are polygons (represented by strings) or textured areas (represented by a graph grammar with error-correction rules). The recognition process is done by a graph matching process that uses a string edit distance to recognize the static parts of the symbol and a parsing process that segments the subgraph in the original graph, following the rules of the graph grammar.
Type de document :
Communication dans un congrès
4th IAPR International Workshop on Graphics Recognition, 2001, Kingston, Ontario, Canada, pp.135-146, 2001
Liste complète des métadonnées

https://hal.inria.fr/inria-00101096
Contributeur : Publications Loria <>
Soumis le : mardi 26 septembre 2006 - 14:56:28
Dernière modification le : jeudi 11 janvier 2018 - 06:19:48

Identifiants

  • HAL Id : inria-00101096, version 1

Collections

Citation

Gemma Sánchez, Josep Lladós, Karl Tombre. An Error-Correction Graph Grammar to Recognize Textured Symbols. 4th IAPR International Workshop on Graphics Recognition, 2001, Kingston, Ontario, Canada, pp.135-146, 2001. 〈inria-00101096〉

Partager

Métriques

Consultations de la notice

74