Abstract
This paper proposed an end-to-end model for recognizing offline handwritten Mongolian words. To be specific, a sequence to sequence architecture with attention mechanism is used to perform the task of generating a target sequence from a source sequence. The proposed model consists of two LSTMs and one attention network. The first LSTM is an encoder which consumes a frame sequence of one word image. The second LSTM is a decoder which can generate a sequence of letters. The attention network is added between encoder and decoder, which allow the decoder to focus on different positions in a sequence of frames during the procedure of decoding. In this study, we have attempted two schemes for generating frames from word images. In the first scheme, frames are generated with overlapping. Each adjacent two frames overlap half a frame. In the second scheme, frames are generated without overlapping. In addition, the height of the frame is also taken into consideration in our study. By comparison, the better scheme for generating frames has been determined. Experimental results demonstrate that the proposed end-to-end model outperforms the state-of-the-art method.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Wei, H., Gao, G.: A keyword retrieval system for historical Mongolian document images. Int. J. Doc. Anal. Recogn. (IJDAR) 17(1), 33–45 (2014)
Wei, H., Gao, G.: Machine-printed traditional Mongolian characters recognition using BP neural networks. In: Proceeding of 2009 International Conference on Computational Intelligence and Software Engineering (CiSE 2009), pp. 1–7. IEEE (2009)
Hu, H., Wei, H., Liu, Z.: The CNN based machine-printed traditional Mongolian characters recognition. In: Proceedings of the 36th Chinese Control Conference (CCC 2017), pp. 3937–3941. IEEE (2017)
Gao, G., Su, X., Wei, H., Gong, Y.: Classical Mongolian words recognition in historical document. In: Proceedings of the 11th International Conference on Document Analysis and Recognition (ICDAR 2011), pp. 692–697. IEEE (2011)
Su, X., Gao, G., Wang, W., Bao, F., Wei, H.: Character segmentation for classical Mongolian words in historical documents. In: Li, S., Liu, C., Wang, Y. (eds.) CCPR 2014, Part II. CCIS, vol. 484, pp. 464–473. Springer, Heidelberg (2014). https://doi.org/10.1007/978-3-662-45643-9_49
Su, X., Gao, G., Wei, H., Bao, F.: A knowledge-based recognition system for historical Mongolian documents. Int. J. Doc. Anal. Recogn. (IJDAR) 19(4), 221–235 (2016)
Yuan, A., Bai, G., Yang, P., Guo, Y., Zhao, X.: Handwritten English word recognition based on convolutional neural networks. In: Proceedings of the 13th International Conference on Frontiers in Handwriting Recognition (ICFHR 2012), pp. 207–212. IEEE (2012)
Yang, W., Jin, L., Tao, D., Xie, Z., Feng, Z.: DropSample: a new training method to enhance deep convolutional neural networks for large-scale unconstrained handwritten Chinese character recognition. Pattern Recogn. 58, 190–203 (2016)
Elleuch, M., Tagougui, N., Kherallah, M.: Towards unsupervised learning for Arabic handwritten recognition using deep architectures. In: Arik, S., Huang, T., Lai, W.K., Liu, Q. (eds.) ICONIP 2015, Part I. LNCS, vol. 9489, pp. 363–372. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-26532-2_40
Kim, I., Xie, X.: Handwritten Hangul recognition using deep convolutional neural networks. Int. J. Doc. Anal. Recogn. (IJDAR) 18(1), 1–13 (2015)
Adak, C., Chaudhuri, B.B., Blumenstein, M.: Offline cursive Bengali word recognition using CNNs with a recurrent model. In: Proceedings of the 15th International Conference on Frontiers in Handwriting Recognition (ICFHR 2016), pp. 429–434. IEEE (2016)
Zhang, X., Tan, C.L.: Unconstrained handwritten word recognition based on trigrams using BLSTM. In: Proceedings of the 22nd International Conference on Pattern Recognition (ICPR 2014), pp. 2914–2919. IEEE (2014)
Messina, R., Louradour, J.: Segmentation-free handwritten Chinese text recognition with LSTM-RNN. In: Proceedings of 13th International Conference on Document Analysis and Recognition (ICDAR 2015), pp. 171–175. IEEE (2015)
Voigtlaender, P., Doetsch, P., Ney, H.: Handwriting recognition with large multidimensional long short-term memory recurrent neural networks. In: Proceedings of the 15th International Conference on Frontiers in Handwriting Recognition (ICFHR 2016), pp. 228–233. IEEE (2016)
Fan, D., Gao, G.: DNN-HMM for large vocabulary Mongolian offline handwriting recognition. In: Proceedings of the 15th International Conference on Frontiers in Handwriting Recognition (ICFHR 2016), pp. 72–77. IEEE (2016)
Fan, D., Gao, G., Wu, H.: MHW Mongolian offline handwritten dataset and its application. J. Chin. Inf. Process. 32(1), 89–95 (2018)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
Graves, A., Mohamed, A., Hinton, G.E.: Speech recognition with deep recurrent neural networks. In: Proceedings of the 38th International Conference on Acoustics, Speech and Signal Processing (ICASSP 2013), pp. 6645–6649 (2013)
Acknowledgement
This paper is supported by the National Natural Science Foundation of China under Grant 61463038.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Wei, H., Liu, C., Zhang, H., Bao, F., Gao, G. (2019). End-to-End Model for Offline Handwritten Mongolian Word Recognition. In: Tang, J., Kan, MY., Zhao, D., Li, S., Zan, H. (eds) Natural Language Processing and Chinese Computing. NLPCC 2019. Lecture Notes in Computer Science(), vol 11839. Springer, Cham. https://doi.org/10.1007/978-3-030-32236-6_19
Download citation
DOI: https://doi.org/10.1007/978-3-030-32236-6_19
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-32235-9
Online ISBN: 978-3-030-32236-6
eBook Packages: Computer ScienceComputer Science (R0)
