基于多重规则和路径评价的在线中英文手写识别方法

doi:10.19678/j.issn.1000-3428.0060600

摘要/Abstract

摘要： 手写文本识别方法主要应用于文本输入技术，对人机交互领域的发展起关键作用。针对多数在线输入法无法识别中英文混合手写识别的问题，提出一种在线中英文混合手写文本识别方法。通过对文本笔画进行基于水平相对位置、垂直重叠率、面积重叠率规则的整合以及连笔切分，得到一系列字符片段，同时利用笔画个数、宽高比、中心偏离、平滑度等几何特征和识别置信度，对字符片段进行中英文分类。在此基础上，根据分类结果并结合自然语言模型的路径评价及动态规划搜索算法，分别对候选的中、英文字符片段进行合并处理，得到待识别的中、英文字符序列，并将其分别送入卷积神经网络的中、英文识别模型中，得到手写文本识别结果。实验结果表明，在线手写中英文混合文本识别正确率达93.67%，不仅能切分在线手写中文文本行，而且对包含字符连笔的在线手写中英文文本行也有较好的切分效果。

关键词: 在线手写识别, 中英文混合手写, 中英文分类, 文本行切分, 路径评价

Abstract: Handwritten text recognition is mainly used in text input technology, which plays a key role in the development of human-computer interaction.To address the lack of functionality for Chinese and English mixed handwritten text recognition in most online input methods, an online Chinese and English mixed handwritten text recognition method is proposed.Through the integration of text strokes based on the horizontal relative position, vertical overlap rate, area overlap rate rules, and continuous stroke segmentation, a series of character segments are obtained.In addition, Chinese and English character segments are classified based on the number of strokes, aspect ratio, center deviation, smoothness, and recognition confidence.On this basis, according to the classification results, combined with the path evaluation of the natural-language model and dynamic programming search algorithm, the candidate and English character segments are combined to obtain the Chinese and English character sequences to be recognized, which are, respectively, sent to the Chinese and English recognition models of the Convolutional Neural Network (CNN) to obtain the handwritten text recognition results.The experimental results show that and the recognition accuracy of the online handwritten Chinese and English mixed text is 93.67%, the proposed method can segment online handwritten Chinese text lines as well as online handwritten Chinese and English text lines containing characters.

Key words: online handwriting recognition, mixed Chinese and English handwriting, Chinese and English classification, text line segmentation, path evaluation

中图分类号:

TP391.43

付鹏斌, 刘鹏辉, 杨惠荣, 董澳静. 基于多重规则和路径评价的在线中英文手写识别方法[J]. 计算机工程, 2022, 48(3): 253-262.

FU Pengbin, LIU Penghui, YANG Huirong, DONG Aojing. Online Chinese and English Handwriting Recognition Method Based on Multiple Rules and Path Evaluation[J]. Computer Engineering, 2022, 48(3): 253-262.

https://www.ecice06.com/CN/Y2022/V48/I3/253

图/表 21

20220401123432

20220401123436

20220401123439

20220401123443

20220401123446

20220401123449

20220401123454

20220401123457

20220401123500

20220401123503

20220401123507

20220401123510

20220401123514

20220401123517

20220401123521

20220401123525

20220401123530

20220401123533

20220401123538

20220401123542

20220401123546

参考文献

[1] 王恺, 王庆人.中英文混合文章识别问题[J].软件学报, 2005, 16(5):149-161. WANG K, WANG Q R.Identification of mixed articles in chinese and english[J].Journal of Software, 2005, 16(5):149-161.(in Chinese)
[2] 任荣梓, 高航.基于反馈合并的中英文混排版面OCR技术研究[J].计算机技术与发展, 2017, 27(3):39-43. REN R Z, GAO H.Research on OCR technology of Chinese and English mixed layout based on feedback merging[J].Computer Technology and Development, 2017, 27(3):39-43.(in Chinese)
[3] WANG D H, LIU C L, ZHOU X D.An approach for real-time recognition of online Chinese handwritten sentences[J].Pattern Recognition, 2012, 45(10):3661-3675.
[4] ZHOU X D, WANG D H, TIAN F, et al.Handwritten Chinese/Japanese text recognition using Semi-Markov conditional random fields[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(10):2413-2426.
[5] CHEN K, TIAN L, DING H, et al.A compact CNN-DBLSTM based character model for online handwritten Chinese text recognition[C]//Proceedings of the 14th IAPR International Conference on Document Analysis and Recognition.Washington D.C., USA:IEEE Press, 2017:1068-1073.
[6] KEYSERS D, DESELAERS T, ROWLEY H A, et al.Multi-language online handwriting recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1180-1194.
[7] CARBUNE V, GONNET P, DESELAERS T, et al.Fast multi-language LSTM-based online handwriting recognition[J].International Journal on Document Analysis and Recognition, 2020, 23(6):89-102.
[8] KANG L, RIBA P, VILLEGAS M, et al.Candidate fusion:integrating language modelling into a sequence-to-sequence handwritten word recognition architecture[J].Pattern Recognition, 2020, 112(6):107-113.
[9] SIMAYI W, IBRAHIM M, ZHANG X Y, et al.A benchmark for unconstrained online handwritten Uyghur word recognition[J].Document Analysis and Recognition, 2020, 23(3):205-218.
[10] YAAGOUP K M M, MUSTAFA M E.Online arabic handwriting characters recognition using deep learning[EB/OL].[2020-12-10].https://www.researchgate.net/publication/330076104_IJARCCE.
[11] GHOSH S, CHATTERJEE A, SEN S, et al.CTRL-CapTuRedLight:a novel feature descriptor for online Assamese numeral recognition[J].Multimedia Tools and Applications, 2021, 80(11):30033-30056.
[12] YAO Z, DING X, LIU C.On-line handwritten Chinese word recognition based on lexicon[C]//Proceedings of the 18th International Conference on Pattern Recognition.Washington D.C., USA:IEEE Press, 2006:320-323.
[13] BAI Z, HUO Q.A study on the use of 8-directional features for online handwritten chinese character recognition[C]//Proceedings of the 8th International Conference on Document Analysis and Recognition.Washington D.C., USA:IEEE Press, 2006:262-266.
[14] WANG Q F.Handwritten Chinese text recognition by integrating multiple contexts[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(8):1469-1481.
[15] 施恩, 李骞, 顾大权等.基于局部特征的卷积神经网络模型[J].计算机工程, 2018, 44(2):282-286. SHI E, LI Q, GU D Q.Convolutional neural network model based on local feature[J].Computer Engineering, 2018, 44(2):282-286.(in Chinese)
[16] GUPTA D, BAG S.CNN-based multilingual handwritten numeral recognition:a fusion-free approach[J].Expert Systems With Applications, 2021, 165:113784-113792.
[17] LAN W, WU L, JIA X Y, et al.Research on application of an improved deep convolutional neural network in handwritten character recognition[J].Journal of Physics.Conference Series, 2020, 1629(1):12002-12013.
[18] YEKTA S C, ERDEM K.Automatic CNN-based arabic numeral spotting and handwritten digit recognition by using deep transfer learning in ottoman population registers[J].Applied Sciences, 2020, 10(16):5430-5441.
[19] LECUN Y, JACKEL L D, BOTTOU L, et al.Comparison of learning algorithms for handwritten digit recognition[C]//Proceedings of International Conference on Artificial Neural Networks.Washington D.C., USA:IEEE Press, 1995, 53-60.
[20] ZHOU S, CHEN H, et al.HIT-OR3C:an opening recognition corpus for Chinese characters[C]//Proceedings of the 9th IAPR International Workshop on Document Analysis Systems.Boston MA, US:[s.n.], 2010:223-230.
[21] JIN L, GAO Y, LIU G, et al.SCUT-COUCH2009-a comprehensive online unconstrained Chinese handwriting database and benchmark evaluation[J].International Journal on Document Analysis and Recognition, 2011, 14(1):53-64.
[22] WANG D H, LIU C L, YU J L, et al.CASIA-OLHWDB1:adatabase of online handwritten Chinese characters[C]//Proceedings of the 10th International Conference on Document Analysis and Recognition.New York, USA:ACM Press, 2009:1206-1210.
[23] WU C, FAN W, HE Y, et al.1st place in ICDAR 2013 chinese handwriting recognition competition[C]//Proceedings of Iapr International Conference on Document Analysis and Recognition.Washington D.C., USA:IEEE Press, 2013:1464-1470.
[24] 陈肇欣.联机手写汉字文本行识别算法研究[D].广州:华南理工大学, 2013. CHEN Z X.Research on on-line handwritten Chinese character text line recognition algorithm[D].Guangzhou:South China University of Technology, 2013.(in Chinese)
[25] 邱理权.基于多种上下文信息的联机手写中文文本识别方法及系统实现[D].广州:华南理工大学, 2017. QU L Q.Multi-contexts based online handwritten Chinese text recognition methods and system implementation[D].Guangzhou:South China University of Technology, 2017.(in Chinese)

选择文件类型/文献管理软件名称

选择包含的内容