全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

小写金额的多模式切分与识别算法

DOI: 10.11834/jig.20080417

Keywords: 多模式切分,识别,小写金额,支票

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对带表格的中文支票小写金额的自动识别问题,提出了一种多模式切分和识别算法。根据小写金额不同部分的切分和识别难度,采取了3种递进的模式:预切分模式、连写0检测模式和基于识别的切分模式。其中预切分模式用来处理小写金额中不粘连的单字;连写0检测模式用来检测并识别连写的0;基于识别的切分模式用来处理非连写0的粘连部分,在这个模式中采用了遗传算法来加速最优解的搜索过程。利用从银行采集的1053张真实支票样本进行测试,在拒识率为33?6%时,小写金额串的识别率达到66?1%,实验结果证明这种算法可以提高真实支票小写金额的识别率。

References

[1]  Nikolai Gorski,Valery Auisimov,Emmanuel Augustin,et al.Sergey maximov industrial bank check processing:the A2iA CheckReaderTM[J].International Journal on Document Analysis and Recognition,2001,3(4):196~206.
[2]  Gu Jun-xia,Ding Xiao-qing.Fusion recognition of courtesy and legs] amounts on Chinese handwritten bank checks[A].In:Proceedings of the International Conference on Signal Processing[C],Guilin,China,2006:1738~1741.
[3]  Yi-Hong Tseng,Hsi-Jian Lee.Interfered-chamcter recognition by removing interfering-lines and adjusting feature weights[A].In:Proceedings of the fourteenth International Conference on Pattern Recognition[C],Brisbane,Austrialia,1998:1865~1867.
[4]  Lei Yun,Lin C S,Din8 X Q,et al.A recngnition based system forsegmentation of touching handwritten numeral strings[A].In:Proceedings of the ninth International Workshop on Frontiers in Handwriting Recognition[C],Tokyo,Japan,2004:294~299.
[5]  Qian Wei-wei,The Back Check Recognition System[D].Beijing:Tsinghua University,2002.[钱伟威.支票识别系统[D].北京:清华大学,2000.]
[6]  Zhang L Q,Suen C Y.Recognition of courtesy amounts on bank checks based on a segmentation approach[A].In:Proceedings of the Eighth International Workshop on Frontiers in Handwriting Recogni-tion[C],Ontario,Canada,2002:298~302.
[7]  Song Hong-dong,Wang Ya-li,Xin Shao-wei.Recognition system of handwritten numerals on checks using support vector machine[J].Computer Engineering and Applications,2003,(3):58~60.[宋鸿冬,王亚利,夏绍玮.基于支撑向量机的支票手写体数字识别系统[J].计算机工程与应用.2003,(3):58~60.]
[8]  Dzuba G,Filatov A,Gershuny D,et al.Chock amount recognition based on the cross validation of courtesy and legal amount fields[J].International Journal on Pattern Recognition and Artificial Intelli-gence,1997,11(4):639~655.
[9]  Tae-Chang Jee,Eun-Jin Kim,Yillbyung Lee.Error correction of Kore-an courtesy amount in hack slips using rule information and cross-ref-erencing[A].In:proceedings of the International Conference on Doc-ument Analysis and Recognition[C],Bangalore,India,1999:95~98.
[10]  Garris M D.Component-Based Handprint Segmentation Using Adaptive Writing Style Model[R].NIST Internal Report 5843,Gaithcrsburg,USA:National Institute of Standards and Technolngy,1996.
[11]  Liu Hai-long,Ding Xino-qing.Handwritten character recngnition using gradient feature and quadratic classifier with multiple discrimination schemes[A].In:Proceedings of the International Conference on Document Analysis and Recngnitiou[C],sooul,Korea,2005:19~23.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133