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...

基于笔划组合的手写数字切分

DOI: 10.11834/jig.20090822

Keywords: 主曲线,分类器组合,置信变换,后验概率,数字串分割,笔划组合

Full-Text   Cite this paper   Add to My Lib

Abstract:

主曲线是一种新的基于非线性变换的特征抽取方法,它是一种通过数据分布“中间”并满足“自相合”的光滑曲线来进行特征提取的方法。为了提高手写数字串切分的正确率,提出了一种基于笔划组合的手写数字串切分方法。该方法首先使用主曲线完成字符模板的笔划抽取,然后以字符识别器提供的置信度为依据来组合笔划,以实现手写数字串的切分过程。另外,在字符识别器设计方面,则是使用基于数字轮廓分段特征与规范化模板特征这两个单特征分类器组合。实验表明,分别基于这两个特征的分类器具有较强的互补性。由于字符识别器的置信度难以真实反映识别结果,为此需使用类条件置信变换法,通过估计分类器的后验概率来对识别器的置信度进行修正。实验结果表明,该方法对于手写数字的分割是有效的。

References

[1]  Lu Yi,Shridhar M,Character segmentation in handwritten words-an overview,Pattern Recognition,1996(1).
[2]  Casey R,Lecolinet E,A survey of methods and strategies in character segmentation,IEEE Transactions on Pattern Analysis and Machine Intelligence,1996(7).
[3]  Chen Y K,Wang J F,Segmentation of single-or multiple-touching handwritten numeral string using background and forground analysis,IEEE Transactions on Pattern Analysis Machine Intelligence,2000(11).
[4]  高庆吉.王晓华.赵为平 对粘连和缺损数字串分割的研究 [J].-模式识别与人工智能2000(1)
[5]  Arica N,Yarman-Vural F T,Optical character recognition for cursive handwriting,IEEE Transactions on Pattern Analysis Machine Intelligence,2002(6).
[6]  Pal U,Belaid A,Choisy C,Touching numeral segmentation using water reservoir concept,Pattern Recognition Letters,2003(1-3).
[7]  Chuang C T,Tseng L Y,A stroke extraction method for multi-font Chinese characters based on the reduced special interval graph,IEEE Transactions on Systems,Man and Cybernetics,1995(7).
[8]  Wang A B,Fan K C,Wu W H,Recursive hierarchical extraction for handwritten Chinese characters,Pattern Recognition,1997(7).
[9]  Liu X,Jia Y,Character stroke extraction based on B-spline curve matching by constrained alternating optimization,Document Analysis and Recognition,2007(1).
[10]  Chiu H P,Tseng D C,A novel stroke-based feature extraction for handwritten Chinese character recognition,Pattern Recognition,1999(12).
[11]  Kim J W,Kim K I,Choi B J,Decomposition of Chinese characters into strokes using mathematical morphology,Pattern Recognition Letters,1999(3).
[12]  Kegl B,Krzyzak A,Piecewise linear skeletonization using principal curves,IEEE Transactions on Pattern Analysis and Machine Intelligence,2002(1).
[13]  娄震.胡钟山.杨静宇.刘克.孙靖夷 基于轮廓分段特征的手写体阿拉伯数字识别 [J].-计算机学报1999(10)
[14]  张重阳.娄震.杨静宇 基于轮廓和统计特征的手写体数字识别 [J].-计算机工程与应用2004(9)
[15]  Smith S J,Bourgoin M O,Sims K,Handwritten character classification using nearest neighbor in large databases,IEEE Transactions on Pattern Analysis and Machine Intelligence,1994(9).
[16]  Lin Xiao-fan,Ding Xiao-qing,Chen Ming,Adaptive confidence transform based classifier combination for Chinese character recognition,Pattern Recognition Letters,1998(10).
[17]  娄震.金忠.杨静宇 基于类条件置信变换的后验概率估计方法 [J].-计算机学报2005(1)
[18]  Kittle J,Hatef M,Duin R P W,On combining classifiers,IEEE Transactions on Pattern Analusis and Machine Intelligence,1998(3).

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133