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...

用于相似字识别的手写汉字特征优化方法

DOI: 10.3969/j.issn.1006-7043.201202024

Keywords: 字符识别, 特征优化, 2DLDA, 手写相似汉字识别

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对常用的手写汉字特征提取方法不利于后续线性区分分析(LDA)特征变换中发现相似汉字的细微区分信息,通过将传统的手写汉字特征提取和LDA变换表述为像素级特征的二维特征矩阵优化问题,并利用二维线性区分分析(2DLDA)变换进行手写汉字特征矩阵的优化,提出了一种用于手写相似汉字识别的特征优化方法.该方法可以避免高维像素级特征向量利用LDA变换进行优化中的散度矩阵奇异性问题.对手写相似汉字的识别实验表明,相对于传统的方法,经过所提方法优化的梯度特征,识别错误率可以降低48.86%,验证了方法的有效性.

References

[1]  LEE S W, PARK J S. Nonlinear shape normalization methods for the recognition of large set handwritten character [J]. Pattern Recognition, 1994, 27(7):895-902.?
[2]  ZHANG H, GUO J, CHEN G, et al. HCL2000―a large-scale handwritten Chinese character database for handwritten character recognition[C]//Proc of The 10th Int Conf Document Analysis and Recognition. Barcelona, Spain, 2009: 286-290.?
[3]  LIU C L, FUJISAWA H. Classification and learning methods for character recognition: advances and remaining problems [M]// MARINAI S, FUJISAWA H. Machine Learning in Document Analysis and Recognition. Berlin: Springer -Verlag, 2008: 139-161.?
[4]  LIU C L, YIN F, WANG D H, et al. Chinese handwriting recognition contest 2010[C] //Proc of 2010 Chinese Conference on Pattern Recognition. Beijing, China, 2010: 1-5.?
[5]  JIN L W, WEI G. Handwritten Chinese character recognition with directional decomposition cellular features [J]. Journal of Circuits, System, and Computers, 1998, 8(4):517-524.?
[6]  DING K, LIU Z B, JIN L W, et al. A comparative study of Gabor feature and gradient feature for handwritten Chinese character recognition[C]//Proc of 2007 Int Conf on Wavelet Analysis and Pattern Recognition. Beijing, China, 2007: 1182-1186.?
[7]  HUO Q, GE Y, FENG Z D. High performance Chinese OCR based on Gabor features, discriminative feature extraction and model training[C]//Proc of IEEE Int Conf Acoustics, Speech, and Signal Processing. Salt Lake City, 2001: 1517-1520.?
[8]  LIU C L, NAKASHIMA K, SAKO H. et al. Handwritten digit recognition: investigation of normalization and feature extraction techniques [J]. Pattern Recognition, 2004, 37(2): 265-279.?
[9]  LIU H, DING X. Handwritten character recognition using gradient feature and quadratic classifier with multiple discrimination schemes [C]//Proc of 8th Int Conf on Document Analysis and Recognition. Seoul, 2005:19-23.?
[10]  BAI Z L, HUO Q. A study on the use of 8-directional features for online handwritten Chinese character recognition[C]//Proc of 8th Int Conf on Document Analysis and Recognition. Seoul, 2005: 262-266.?
[11]  LIU C L. Normalization-cooperated gradient feature extraction for handwritten character recognition [J]. IEEE Trans Pattern Analysis and Machine Intelligence, 2007. 29(8):1465-1469.?
[12]  LONG T, JIN L W. Building compact MQDF classifier for large character set recognition by subspace distribution sharing [J]. Pattern Recognition, 2008, 41(9): 2916-2925.?
[13]  FISHER R A. The use of multiple measurements in taxonomic problems [J]. Annals of Eugenics, 1936, 7: 179-188.?
[14]  MING L, YUAN B. 2D-LDA: a statistical linear discriminant analysis for image matrix [J]. Pattern Recognition Letters, 2005, 26(5): 527-532.?
[15]  NOUSHATH S, HEMANTHA K G, SHIVAKUMARA P. (2D)2LDA: an efficient approach for face recognition [J]. Pattern Recognition, 2006, 39(7):1396-1400.?
[16]  GAO X, WEN W H, JIN L W. A new feature optimization method based on two-directional 2DLDA for handwritten Chinese character recognition [C]//Proc of the 11th Int Conf Document Analysis and Recognition. Beijing, China, 2011:232-236.?
[17]  RAO C R. The utilization of multiple measurements in problems of biological classification [J]. J Royal Statistical Soc. B: Methodological, 1948, 10: 159-203.?
[18]  YE J, JANARDAN R, LI Q. Two-dimensional linear discriminant analysis [C] //Proc of the 8th Annual Conf on Neural Information Processing Systems. Victoria, British Columbia, Canada, 2004: 1569-1576.?
[19]  YANG J, ZHANG D, YONG X, et al. Two-dimensional discriminant transform for face recognition [J]. Pattern Recognition, 2005, 38(7):1125-1129.?

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133