Strategies for combining on-line and off-line information in an on-line handwriting recognition system

被引:9
作者
Gauthier, N [1 ]
Artières, T [1 ]
Dorizzi, B [1 ]
Gallinari, P [1 ]
机构
[1] Univ Paris 06, F-75015 Paris, France
来源
SIXTH INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION, PROCEEDINGS | 2001年
关键词
D O I
10.1109/ICDAR.2001.953823
中图分类号
TP18 [人工智能理论];
学科分类号
081104 [模式识别与智能系统]; 0812 [计算机科学与技术]; 0835 [软件工程]; 1405 [智能科学与技术];
摘要
This paper investigates the cooperation of on-line and off-line handwriting word recognition systems. Our goal is to improve a mature on-line recognition system by exploiting the complementary information present in the off-line representation built from on-line signal. After describing the on-line and off-line HMM based handwriting recognition systems. we propose a formal framework. which allows describing different strategies for combining the two HMMs. These schemes are then evaluated on the UNIPEN database, both for isolated character and word recognition tasks.
引用
收藏
页码:412 / 416
页数:3
相关论文
共 12 条
[1]
BRAKENSIEK A, 1999, INT C DOC AN REC ICD
[2]
DUPONT S, 1998, INT C SPOK LANG PROC
[3]
GARCIASALICETTI S, 1995, HIDDEN MARKOV MODEL
[4]
GUYON I, 1994, INT C PATT REC ICPR
[5]
Writer independent on-line handwriting recognition using an HMM approach [J].
Hu, JY ;
Lim, SG ;
Brown, MK .
PATTERN RECOGNITION, 2000, 33 (01) :133-147
[6]
THE SEGMENTAL K-MEANS ALGORITHM FOR ESTIMATING PARAMETERS OF HIDDEN MARKOV-MODELS [J].
JUANG, BH ;
RABINER, LR .
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1990, 38 (09) :1639-1641
[7]
MANKE S, 1995, NPENPLUSPLUS WRITER
[8]
MARTI UV, 2000, HANDWRITTEN SENTENCE
[9]
NEY H, 1996, AUTOMATIC SPEECH SPE, P51
[10]
SEILER R, 1996, OFF LINE CURSIVE HAN