Abstract
There are vast amount of historical documents written in cursive writing style in Japan. However, characters written by the style cannot be read by modern people, because the style was taught no longer. Therefore, an efficient method to convert historical characters into modern characters automatically is required. Especially, since every page in a Japanese historical document is stored by a photo image, it is necessary to automatically recognize all characters in a photo image. However, it is difficult to recognize each historical characters separately, because they are written connected and because there are many types of shape of characters. In this paper, we propose a method combining a method using deep learning to detect characters in a photo image and a method separating a block into characters. The remained parts that cannot be recognized by the former method are separated into characters by the latter method. Thus, it is expected that the recognition ratio is improved. We evaluate the performance of the proposed algorithm by using photo images of actual documents.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Shoten, I.: General Catalog of National Books. Iwanami Shoten (2002)
Shibayama, M., et. al.: Research on Higher Accuracy Document Character Recognition Systems (in Japanese), Grants-in-Aid for Scientific Research (B) (1) report on research results, pp. 33–49 (2005)
Yamamoto, S., Osawa, T.: Labor saving for reprinting Japanese rare classical books: the development of the new method for OCR technology including Kana and Kanji characters in cursive style. Inf. Manag. 58(11), 819–826 (2016)
Hayasaka, T., Ohno, W., Kato, Y., Yamamoto, K.: Recognition of Hentaigana by deep learning and trial production of WWW application (in Japanese). In: Proceedings of IPSJ Symposium of Humanities and Computer Symposium, pp. 7–12 (2016)
The Dataset of Kuzushiji (Open Data Center for Humanities). http://codh.rois.ac.jp/pmjt
Tarin, C., Mikel, B., Asanobu, K., Alex, L., Kazuaki, Y., David, H.: Deep learning for classical Japanese literature. In: Proceedings of 2018 Workshop on Machine Learning for Creativity and Design (Thirty-second Conference on Neural Information Processing Systems), 3 December 2018
Nguyen, H.T., Ly, N.T., Nguyen, K.C., Nguyen, C.T., Nakagawa, M.: Attempts to recognize anomalously deformed Kana in Japanese historical documents. In: Proceedings of 4th International Workshop on Historical Document Imaging and Processing, pp. 31–36, 10–11 November 2017
Clanuwat, T., Alex, L., Asanobu, K.: End-to-end pre-modern Japanese character (Kuzushiji) spotting with deep learning. In: Proceedings of IPSJ SIG Computers and the Humanities, pp. 15–20 (2018)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Medical Image Computing and Computer-Assisted Intervention (MICCAI). LNCS, vol. 9351, pp. 234–241. Springer (2015)
Sichao, L., Miwa, H.: Algorithm using deep learning for recognition of Japanese historical characters in photo image of historical book. In: Proceedings of INCoS, Oita, Japan, pp. 5–7, September 2019
Kingama, D.P., Ba, J.L.: A method for stochastic optimization. In: Proceedings of ICLR, San Diego, 7–9 May 2015
Wu, Y., He, K.: Group normalization. In: Proceedings of ECCV, Munich, Germany, 8–14 September 2018
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: Proceedings of International Conference on Learning Representations, San Diego, 7–9 May 2015
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Sichao, L., Miwa, H. (2020). Recognition of Historical Characters by Combination of Method Detecting Character in Photo Image of Document and Method Separating Block to Characters. In: Barolli, L., Okada, Y., Amato, F. (eds) Advances in Internet, Data and Web Technologies. EIDWT 2020. Lecture Notes on Data Engineering and Communications Technologies, vol 47. Springer, Cham. https://doi.org/10.1007/978-3-030-39746-3_48
Download citation
DOI: https://doi.org/10.1007/978-3-030-39746-3_48
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-39745-6
Online ISBN: 978-3-030-39746-3
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)