Rejection strategies for offline handwritten sentence recognition

Author

Zimmermann, Matthias ; Bertolami, Roman ; Bunke, Horst

Author_Institution

Inst. of Informatics & Appl. Mathematics, Bern Univ., Switzerland

Volume

2

fYear

2004

fDate

23-26 Aug. 2004

Firstpage

550

Abstract

This work investigates three different rejection strategies for offline handwritten sentence recognition. The rejection strategies are implemented as a postprocessing step of a hidden Markov model based text recognition system and are based on confidence measures derived from a list of candidate sentences produced by the recognizer. The better performing confidence measures make use of the fact that the recognizer integrates a word bigram language model. Experimental results on extracted sentences from the IAM database validate the effectiveness of the proposed rejection strategies.

Keywords

handwritten character recognition; hidden Markov models; confidence measures; hidden Markov model based text recognition system; offline handwritten sentence recognition; rejection strategies; word bigram language model; Artificial neural networks; Character recognition; Databases; Handwriting recognition; Hidden Markov models; Informatics; Mathematics; Performance evaluation; Speech recognition; Text recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on

ISSN

1051-4651

Print_ISBN

0-7695-2128-2

Type

conf

DOI

10.1109/ICPR.2004.1334301

Filename

1334301