Dedicated Texture Based Tools for Characterisation of Old Books

Abstract : This paper deals with the development of suitable assistance tools for humanists and historians to help them retrieve information documents. This paper represents a part of this ambitious project and deals with the design of a pixel classification method for ancient typewritten documents. The approach presented here is based on a multiresolution map construction and analysis; for five resolutions we construct five different characterisation maps. All the maps are based on texture information (correlation of pixels orientations, grey level pixel density: etc). After the merging of these 25 maps, each pixel of the original image is described by a vector which allows the computing of a hierarchical classification. In order to avoid issues linked to the binarization process, ail maps are computed for grey level images. The system has been tested on a CESR database of ancient printed books of the Renaissance. The classification results are evaluated through several visual classification illustrations
Type de document :
Communication dans un congrès
Second International Conference on Document Image Analysis for Libraries, 2006, Apr 2006, Lyon, France. IEEE, pp.60-69, 2006, 〈10.1109/DIAL.2006.11〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00355268
Contributeur : Nicholas Journet <>
Soumis le : jeudi 22 janvier 2009 - 14:43:18
Dernière modification le : jeudi 16 mars 2017 - 01:05:30

Identifiants

Collections

Citation

Nicholas Journet, Véronique Eglin, Rémy Mullot, Jean-Yves Ramel. Dedicated Texture Based Tools for Characterisation of Old Books. Second International Conference on Document Image Analysis for Libraries, 2006, Apr 2006, Lyon, France. IEEE, pp.60-69, 2006, 〈10.1109/DIAL.2006.11〉. 〈hal-00355268〉

Partager

Métriques

Consultations de la notice

161