Frequent Graph Discovery: Application to Line Drawing Document Images

Abstract : In this paper a sequence of steps is applied to a graph representation of line drawings using concepts from data mining. This process finds frequent subgraphs and then association rules between these subgraphs. The distant aim is the automatic discovery of symbols and their relations, which are parts of the document model. The main outcome of our work is firstly an algorithm that finds frequent subgraphs in a single graph setting and secondly a modality to find rules and meta-rules between the discovered subgraphs. The searched structures are closed [1] and disjunct subgraphs. One aim of this study is to use the discovered symbols for classification and indexation of document images when a supervised approach is not at hand. The relations found between symbols can be used in segmentation of noisy and occluded document images. The results show that this approach is suitable for patterns, symbols or relation discovery.
Type de document :
Article dans une revue
Electronic Letters on Computer Vision and Image Analysis, ELCVIA, 2005, 5 (2), pp 47-57
Liste complète des métadonnées

Littérature citée [17 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-00440025
Contributeur : Pierre Héroux <>
Soumis le : mercredi 9 décembre 2009 - 10:46:13
Dernière modification le : mercredi 11 octobre 2017 - 11:18:03
Document(s) archivé(s) le : samedi 26 novembre 2016 - 15:14:41

Fichiers

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

Identifiants

  • HAL Id : hal-00440025, version 1

Collections

Citation

Eugen Barbu, Pierre Héroux, Sébastien Adam, Eric Trupin. Frequent Graph Discovery: Application to Line Drawing Document Images. Electronic Letters on Computer Vision and Image Analysis, ELCVIA, 2005, 5 (2), pp 47-57. 〈hal-00440025〉

Partager

Métriques

Consultations de
la notice

124

Téléchargements du document

75