格式化约束的时间戳文字识别网络

doi:10.3778/j.issn.1002-8331.2106-0332

摘要/Abstract

摘要： 文字识别是深度学习网络的重要应用领域，主流算法基于光学信息预测自然场景文字。然而在一些特定领域的文本对象上，额外的关键特征将会进一步提高文字识别算法的准确性。在安防监控领域，画面中的时间戳文本拥有格式规范、限定数值范围等特点，根据这一特点，对时间戳文本识别网络进行了研究，提出一种时间戳信息约束机制，融合文本语义约束信息和光学特征达到识别规范文本的效果，增强输出时间戳文本的格式规范性和数值合理性。在全匹配率、编辑距离等标准上全面超过基于光学特征的经典文字识别算法。

关键词: 文字识别, 图像语义信息, 智能监控, 时间戳文本, 格式化文本识别

Abstract: Text recognition is an important application field of in-depth learning network. Mainly used algorithms predict text in natural scenes based on optical information. However, on some domain-specific text objects, additional key features will further improve the accuracy of text recognition algorithms. In the field of security monitoring, timestamp text in pictures has the characteristics of standard format and limited value range. This paper studies the timestamp text recognition network for this feature, and puts forward a time stamp information constraint mechanism, which combines the text semantic constraint information with the optical characteristics to achieve the effect of identifying standard text, and enhance formatting specifications and numerical rationality of the output timestamp text. Classic text recognition algorithms based on optical features are completely surpassed in terms of full match rate and editing distance.

Key words: text recognition, image semantics, intelligent monitoring, timestamp, formated text recognition

刘洋, 陈黎. 格式化约束的时间戳文字识别网络[J]. 计算机工程与应用, 2022, 58(24): 189-195.

LIU Yang, CHEN Li. Time Stamp Recognition Network Constrainted by Formated Information[J]. Computer Engineering and Applications, 2022, 58(24): 189-195.

参考文献

[1] RUSSELL S，NORVIG P.Artificial intelligence：a modern approach[M].[S.l.]：Prentice Hall，1995.
[2] NIBLACK W.An introduction to digital image processing[M].USA：Prentice Hall，1986.
[3] SMITH R.An overview of the tesseract OCR engine[C]//Ninth International Conference on Document Analysis and Recognition（ICDAR 2007），Curitiba，Brazil，2007：629-633.
[4] BISSACCO A，CUMMINS M，NETZER Y，et al.PhotoOCR：reading text in uncontrolled conditions[C]//IEEE International Conference on Computer Vision，Sydney，NSW，Australia，2013：785-792.
[5] WANG K，BABENKO B，BELONGIE S.End-to-end scene text recognition[C]//International Conference on Computer Vision，Barcelona，Spain，2011.
[6] JADERBERG M，SIMONYAN K，VEDALDI A，et al.Reading text in the wild with convolutional neural networks[J].International Journal of Computer Vision，2015，116：1-20.
[7] ALMAZáN J，GORDO A，FORNéS A，et al.Word spotting and recognition with embedded attributes[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2014，36（12）：2552-2566.
[8] GORDO A.Supervised mid-level features for word image representation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2015：2956-2964.
[9] GORDO A，ALMAZAN J，MURRAY N，et al.LEWIS：latent embeddings for word images and their semantics[C]//Proceedings of the IEEE International Conference on Computer Vision，2015：1242-1250.
[10] RODRIGUEZ-SERRANO J A，PERRONNIN F.Label embedding for text recognition[C]//British Machine Vision Conference，2013.
[11] SHI B，BAI X，YAO C.An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39（11）：2298-2304.
[12] HE P，HUANG W，QIAO Y，et al.Reading scene text in deep convolutional sequences[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2016，30（1）.
[13] GRAVES A，FERNáNDEZ S，GOMEZ F，et al.Connectionist temporal classification：labelling unsegmented sequence data with recurrent neural networks[C]//Proceedings of the 23rd International Conference on Machine Learning，2006：369-376.
[14] LEE C Y，OSINDERO O.Recursive recurrent nets with attention modeling for ocr in the wild[C]//IEEE Conference on Computer Vision and Pattern Recognition（CVPR），2016：2231-2239.
[15] LI H，WANG P，SHEN C，et al.Show，attend and read：a simple and strong baseline for irregular text recognition[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2019：8610-8617.
[16] SHI B G，YANG M K，WANG X G，et al.ASTER：an attentional scene text recognizer with flexible rectification[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2019，41（9）：2035-2048.
[17] QIAO Z，ZHOU Y，YANG D B，et al.SEED：semantics enhanced encoder-decoder framework for scene text recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2020：13528-13537.
[18] MILLER G A.WordNet：a lexical database for english[J].Communications of the Association for Computing Machinery，1995，38（11）：39-41.
[19] PATEL Y，BIGORDA L G I，RUSI?OL M，et al.Dynamic lexicon generation for natural scene images[C]//Computer Vision ECCV 2016 Workshops，2016：395-410.
[20] DAVID M B，ANDREW Y N，MICHAEL I J.Latent dirichlet allocation[J].Journal of Machine Learning Research，2003，3（1）：993-1022.
[21] KANG C，KIM G，YOO S.Detection and recognition of text embedded in online images via neural context models[C]//Association for the Advancement of Artificial Intelligence，2017.
[22] BOJANOWSKI P，GRAVE E，JOULIN A，et al.Enriching word vectors with subword information[J].arXiv：1607.04606，2016.
[23] WANG T W，ZHU Y Z，JIN L W，et al.Decoupled attention network for text recognition[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2020：12216-12224.