SCIENCE CHINA Information Sciences, Volume 64 , Issue 3 : 139101(2021) https://doi.org/10.1007/s11432-018-9824-9

Robust encoder–decoder learning framework for offline handwritten mathematical expression recognition based on a multi-scale deep neural network

More info
  • ReceivedJul 3, 2018
  • AcceptedFeb 11, 2019
  • PublishedMay 27, 2020


There is no abstract available for this article.


This work was supported by National Key R$\&$D Program of China (Grant No. 2016YFE0204200) and National 1000-Talent Youth Program. The authors want to thank Dr. Jianshu ZHANG for insightful comments and suggestions.


Appendixes A–F.


[1] Anderson R H. Syntax-directed recognition of hand-printed two-dimensional mathematics. In: Proceedings of the Association for Computing Machinery Inc. Symposium. New York: ACM, 1968. 436--459. Google Scholar

[2] Lavirotte S, Pottier L. Mathematical formula recognition using graph grammar. In: Proceedings of SPIE, 1998. 3305: 44--52. Google Scholar

[3] Chan K F, Yeung D Y. Error detection, error correction and performance evaluation in on-line mathematical expression recognition. Pattern Recognition, 2001, 34: 1671-1684 CrossRef Google Scholar

[4] Yamamoto R. On-Line Recognition of Handwritten Mathematical Expression Based on Stroke-Based Stochastic Context-Free Grammar. In: Proceedings of the 10th International Workshop on Frontiers in Handwriting Recognition, La Baule, 2006. Google Scholar

[5] MacLean S, Labahn G. A new approach for recognizing handwritten mathematics using relational grammars and fuzzy sets. IJDAR, 2013, 16: 139-163 CrossRef Google Scholar

[6] Huang G, Liu Z, Weinberger K Q. Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017. 2261--2269. Google Scholar

[7] Cho K, Merrienboer B, Bahdanau D, et al. On the properties of neural machine translation: Encoderdecoder approaches. 2014,. arXiv Google Scholar