A generic approach for OCR performance evaluation

Abdel Belaïd 1 Laurent Pierron 2
1 READ - READ
LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : For different document automation operations it is always needed to have an OCR evaluation phase to select the most interesting OCRs for the document class studied. The evaluation should indicate the defects and drawbacks of each OCR and allow to determine the required heuristics to combine these OCRs in order to obtain the highest performances in production: the lowest reject rate for a predefined confusion rate ( in general 1/10000). The evaluation should be done automatically and completely integrated in a more global OCR platform. In this paper, we present our experience in OCR evaluation for four kind of commercial OCRs on few document classes. We will comment the results, the methodology used, the encountered problems and propose some heuristics to improve these results.
Type de document :
Communication dans un congrès
SPIE. Electronic Imaging, 2002, San Jose, California, 5 p, 2002
Liste complète des métadonnées

https://hal.inria.fr/inria-00100718
Contributeur : Publications Loria <>
Soumis le : mardi 26 septembre 2006 - 14:49:58
Dernière modification le : mardi 24 avril 2018 - 13:32:45

Identifiants

  • HAL Id : inria-00100718, version 1

Collections

Citation

Abdel Belaïd, Laurent Pierron. A generic approach for OCR performance evaluation. SPIE. Electronic Imaging, 2002, San Jose, California, 5 p, 2002. 〈inria-00100718〉

Partager

Métriques

Consultations de la notice

145