Handling out-of-vocabulary words and recognition errors based on word linguistic context for handwritten sentence recognition

Solen Quiniou 1, * Mohamed Cheriet 1 Eric Anquetil 2
* Auteur correspondant
1 ETS - Ecole de Technologie Supérieure [Montréal]
2 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 investigate the use of linguistic information given by language models to deal with word recognition errors on handwritten sentences. We focus especially on errors due to out-of-vocabulary (OOV) words. First, word posterior probabilities are computed and used to detect error hypotheses on output sentences. An SVM classifier allows these errors to be categorized according to defined types. Then, a post-processing step is performed using a language model based on Part-of-Speech (POS) tags which is combined to the n-gram model previously used. Thus, error hypotheses can be further recognized and POS tags can be assigned to the OOV words. Experiments on on-line handwritten sentences show that the proposed approach allows a significant reduction of the word error rate.
Type de document :
Communication dans un congrès
International Conference on Document Analysis and Recognition (ICDAR), Jul 2009, Barcelona, Spain. pp.466-470, 2009
Liste complète des métadonnées

Littérature citée [15 références]  Voir  Masquer  Télécharger
https://hal.archives-ouvertes.fr/hal-00582433
Contributeur : Solen Quiniou <>
Soumis le : vendredi 1 avril 2011 - 14:34:54
Dernière modification le : jeudi 7 février 2019 - 16:51:02
Document(s) archivé(s) le : samedi 2 juillet 2011 - 02:44:52

Fichier

quiniou09handling.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00582433, version 1

Collections

UNIV-RENNES1 | IRISA | INSA-RENNES | IRISA-D6 | UR1-UFR-ISTIC | UNIV-RENNES

Citation

Solen Quiniou, Mohamed Cheriet, Eric Anquetil. Handling out-of-vocabulary words and recognition errors based on word linguistic context for handwritten sentence recognition. International Conference on Document Analysis and Recognition (ICDAR), Jul 2009, Barcelona, Spain. pp.466-470, 2009. 〈hal-00582433〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

242

Téléchargements de fichiers

307

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD