Rejection strategies for offline handwritten text line recognition

被引:21
作者
Bertolami, Roman [1 ]
Zimmermann, Matthias [1 ]
Bunke, Horst [1 ]
机构
[1] Univ Bern, Inst Comp Sci & Appl Math, CH-3012 Bern, Switzerland
基金
瑞士国家科学基金会;
关键词
handwritten text recognition; rejection strategies; statistical language model;
D O I
10.1016/j.patrec.2006.06.002
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper investigates rejection strategies for unconstrained offline handwritten text line recognition. The rejection strategies depend on various confidence measures that are based on alternative word sequences. The alternative word sequences are derived from specific integration of a statistical language model in the hidden Markov model based recognition system. Extensive experiments on the IAM database validate the proposed schemes and show that the novel confidence measures clearly outperform two baseline systems which use normalised likelihoods and local n-best lists, respectively. (c) 2006 Elsevier B.V. All rights reserved.
引用
收藏
页码:2005 / 2012
页数:8
相关论文
共 27 条
[1]   An omnifont open-vocabulary OCR system for English and Arabic [J].
Bazzi, I ;
Schwartz, R ;
Makhoul, J .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1999, 21 (06) :495-504
[2]  
Brakensiek A, 2004, LECT NOTES COMPUT SC, V2956, P103
[3]  
Brakensiek A., 2000, 7 INT WORKSH FRONT H, P343
[4]  
Gorski N, 1997, PROC INT CONF DOC, P1092, DOI 10.1109/ICDAR.1997.620677
[5]  
Jelinek, 1990, READINGS SPEECH RECO, P450, DOI [10.1016/B978-0-08-051584-7.50045-0, DOI 10.1016/B978-0-08-051584-7.50045-0]
[6]  
Johansson S., 1986, TAGGED LOB CORPUS US
[7]   ESTIMATION OF PROBABILITIES FROM SPARSE DATA FOR THE LANGUAGE MODEL COMPONENT OF A SPEECH RECOGNIZER [J].
KATZ, SM .
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1987, 35 (03) :400-401
[8]  
Kim G, 1999, ADV HANDWRITING RECO, P163
[9]   Rejection strategies for handwritten word recognition [J].
Koerich, AL .
NINTH INTERNATIONAL WORKSHOP ON FRONTIERS IN HANDWRITING RECOGNITION, PROCEEDINGS, 2004, :479-484
[10]  
Maltoni D., 2003, HDB FINGERPRINT RECO