Document Image Characterization Using a Multiresolution Analysis of the Texture: Application to Old Document

Abstract : In this article, we propose a method of characterization of images of old documents based on a texture approach. This characterization is carried out with the help of a multi-resolution study of the textures contained in the images of the document. Thus, by extracting five features linked to the frequencies and to the orientations in the different areas of a page, it is possible to extract and compare elements of high semantic level without expressing any hypothesis about the physical or logical structure of the analyzed documents. Experimentation based on segmentation, data analysis and document image retrieval tools demonstrate the performance of our propositions and the advances that they represent in terms of characterization of content of a deeply heterogeneous corpus.
Type de document :
Article dans une revue
International Journal on Document Analysis and Recognition, Springer Verlag, 2008, 11 (1), pp.9-18
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01022532
Contributeur : Denis Maurel <>
Soumis le : jeudi 10 juillet 2014 - 14:48:18
Dernière modification le : jeudi 9 février 2017 - 16:58:52

Identifiants

  • HAL Id : hal-01022532, version 1

Collections

Citation

Nicholas Journet, Jean-Yves Ramel, Rémy Mullot, Véronique Eglin. Document Image Characterization Using a Multiresolution Analysis of the Texture: Application to Old Document. International Journal on Document Analysis and Recognition, Springer Verlag, 2008, 11 (1), pp.9-18. <hal-01022532>

Partager

Métriques

Consultations de la notice

115