一种云平台下高识别率的手写汉字光学图像识别系统

J4 ›› 2016, Vol. 33 ›› Issue (5): 530-536.

一种云平台下高识别率的手写汉字光学图像识别系统

胡晓芳¹,赵元黎²

1长治学院电子信息与物理系，山西长治 046000；2郑州大学物理工程学院，河南郑州 450001

收稿日期:2015-12-16 修回日期:2016-07-22 出版日期:2016-09-28 发布日期:2016-09-28
通讯作者: 赵元黎(1957－) ,女,博士,教授,研究方向:测控计量技术。
作者简介:胡晓芳（1974-），女（汉族），山西省晋中市，硕士，讲师，主要研究领域为光学图像处理、数字信号处理、数字电路.
基金资助:
国家自然科学基金资助项目(10974183)

An optical image recognition system for handwritten Chinese character recognition rate in cloud platform

Hu Xiaofang, Zhao Yuanli

1Department of Electronic Information and Physics，Changzhi University，Changzhi Shanxi 046011,China;2 School of Physics and Engineering,Zhengzhou University,Zhengzhou 450001,China

Received:2015-12-16 Revised:2016-07-22 Published:2016-09-28 Online:2016-09-28

摘要/Abstract

摘要： 针对原有手写汉字识别系统中文字特征提取的相关问题，笔者在本篇文章中设计了一类全新的方法，本方法结合卷积神经网对字形相似的字智能化学习有效特征，并且采用了光学图像识别技术，另外，这类方法通过手写云平台中丰富的数据资源对模型进行高效的训练，同时根据频度统计形成特定的相似子集，以此有效的优化识别率。实验结果显示，和原有的支持向量机(SVM)以及最近邻分类器方法进行系统性的对比，本文所论述的方法能够显著提升识别率。

关键词: 手写汉字；自动学习；卷积神经网；云平台；识别率

Abstract: For the feature extraction methods’ restrictions problem of handwritten chinese character recognition in traditional two handwritten cinese character recognition system, the paper proposes a identification system method which uses convolution neural network to automatically learn chinese characters similar characteristics. The method uses data from large handwritten cloud platform to train the model, generating similar frequency statistics based on a subset of, and further improve the recognition rate. Experimental results show that the recognition rate with respect to the traditional gradient-based feature support vector machine (SV and nearest neighbor classifier method, this method has improved to some extent.

Key words: Handwritten cinese character; Automatic learning; Convolution neural network; Cloud platform; Recognition rate

胡晓芳赵元黎. 一种云平台下高识别率的手写汉字光学图像识别系统[J]. J4, 2016, 33(5): 530-536.

Hu Xiaofang, Zhao Yuanli. An optical image recognition system for handwritten Chinese character recognition rate in cloud platform[J]. J4, 2016, 33(5): 530-536.

参考文献

[1]吴健辉，张国云，杨坤涛，基于置信度的多分类器互补集成手写数字识别[J]. 计算机工程与应用，2007,43（30）：228-230.
[3]廖斌,刘鸳鸳.基于多尺度灰度变换的图像増强研究[J].量子电子学报,2015,32（5）：550-554.
[3]K. Leung, C. Leung, “Recognition of handwritten Chinese characters by critical region analysis” [J]. Pattern Recognition, 2010,43(3): 949-961.
[4]GAO T F, LIU C L. High accuracy handwritten Chinese character recognition using LDA-based compound distances[J]. Pattern Recognition, 2008, 41(11): 3442-3451.
[5]D. Tao, L. Liang, L. Jin, et al, “Similar handwritten Chinese character recognition by kernel discriminative locality alignment”[J], Pattern Recognition Letters, 35(1): 186-194, 2014.
[6]LECUN Y, BOSERB, DENKER J S, et al. Handwritten digit recognition with a back-propagation network[A]. Advances in neural information processing systems[C]. Denver, United States, 1990. 396-404.
[7] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[8] DENG H, STATHOPOULOS G, SUEN C Y. Error-correcting output coding for the convolutional neural network for optical character recognition[A]. Proceedings of IEEE International Conference on Document Analysis and Recognition[C].Barcelona, Spain, 2009: 581-585.
[9] YUAN A, BAI G, JIAO L, et al. Offline handwritten English character recognition based on convolutional neural network[A]. Proceedings of IAPR International Workshop on Document Analysis Systems[C]. Gold Cost, QLD, 2012. 125-129.
[10]NETZER Y, WANG T, COATES A, et al. Reading digits in natural images with unsupervised feature learning[A]. NIPS Workshop on Deep Learning and Unsupervised Feature Learning[C]. Granada, Spain, 2011.
[11] SERMANET P, CHINTALA S, LECUN Y. Convolutional neural networks applied to house numbers digit classification[A].Proceedings of IEEE International Conference on Pattern Recognition[C].Tsukuba, Japan, 2012. 3288-3291.
[12]COATES A, CARPENTER B, CASE C, et al. Text detection and character recognition in scene images with unsupervised feature learning[A]. Proceedings of IEEE International Conference on Document Analysis and Recognition[C]. Beijing, China, 2011. 440-445.
[13]WANG T, WU D J, COATES A, et al. End-to-end text recognition with convolutional neural networks[A]. Proceedings of IEEE International Conference on Pattern Recognition[C].Tsukuba, Japan, 2012. 3304-3308.
[14]SIMARD P, STEINKRAUS D, PLATT J C. Best practices for convolutional neural networks applied to visual document analysis[A]. Proceedings of IEEE International Conference on Document Analysis and Recognition[C]. Edinburgh, UK, 2003.958-963.
[15]CIRESAN D C, MEIER U, GAMBARDELLA L M, et al. Convolutional neural network committees for handwritten character classification[A]. Proceedings of IEEE International Conference on Document Analysis and Recognition[C]. Beijing, China, 2011. 1135-1139.
[16]VAQUERO L M, CACERES J, MORAN D. The challenge of service level scalability for the cloud[J]. International Journal of Cloud Applications and Computing (IJCAC), 2011, 1(1):34-44.
[17]GAO Y, JIN L W, HE C, et al. Handwriting character recognition as a service: a new handwriting recognition system based on cloud computing[A]. Proceedings of IEEE International Conference on Document Analysis and Recognition[C]. Beijing, China, 2011. 885-889.
[18]LIU Z B, JIN L W. A static candidates generation technique and its application in two-stage LDA Chinese character recognition[A]. Proceedings of Chinese Control Conference[C]. Hunan, China, 2007. 571-575.