Student Movement Documents and Related Social Movement Materials in Postwar Japan: A Dataset of Images, Layout Annotations, and Line-Level Transcriptions

Shumpei Katakura

doi:10.5334/johd.629

Student Movement Documents and Related Social Movement Materials in Postwar Japan: A Dataset of Images, Layout Annotations, and Line-Level Transcriptions

Journal of Open Humanities Data

Volume 12 (2026): Issue 1

By: Shumpei Katakura

Open Access

|Sep 2026

Abstract

This paper presents an image–annotation dataset of student movement and related social movement materials held by the Tohoku University Archives. Version 1.1.0 contains 690 distributed JPEG and TIFF images derived from 676 pre-split images and 690 corresponding UTF-8 JSON files, with 5,781 block annotations and 47,631 line annotations. The records include bounding boxes, manual transcriptions, writing direction, and printed/handwritten classifications. Two JSON implementations created under a shared functional specification are preserved. The dataset supports OCR/HTR, layout analysis, page-text reconstruction, document-image understanding, and search-oriented evaluation.

References

Boillet, M., Kermorvant, C., & Paquet, T. (2022). Robust Text Line Detection in Historical Documents: Learning and Evaluation Methods. International Journal on Document Analysis and Recognition, 25, 95–114. 10.1007/s10032-022-00395-7
Open DOI Search in Google Scholar Back to article
Clanuwat, T., Bober-Irizar, M., Kitamoto, A., Lamb, A., Yamamoto, K., & Ha, D. (2018). Deep Learning for Classical Japanese Literature. 10.48550/arXiv.1812.01718
Open DOI Search in Google Scholar Back to article
IIIF Consortium (2020). IIIF Presentation API 3.0. Retrieved from https://iiif.io/api/presentation/3.0/ (Accessed 5 August 2026).
Search in Google Scholar Back to article
Katakura, S., Tsukagoshi, Y., Hanzawa, T., & Yanagihara, S. (2025). Tohoku University Digital Archive: Characteristics of Archived Materials and Prospects for AI Utilization. Journal of Japan Society of Information and Knowledge, 35(2), 227–232. 10.2964/jsik_2025_020
Open DOI Search in Google Scholar Back to article
Martínek, J., Lenc, L., & Král, P. (2020). Building an Efficient OCR System for Historical Documents with Little Training Data. Neural Computing and Applications, 32, 17209–17227. 10.1007/s00521-020-04910-x
Open DOI Search in Google Scholar Back to article
Shen, Z., Zhang, K., & Dell, M. (2020). A Large Dataset of Historical Japanese Documents with Complex Layouts. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (pp. 2336–2343). Retrieved from https://openaccess.thecvf.com/content_CVPRW_2020/papers/w34/Shen_A_Large_Dataset_of_Historical_Japanese_Documents_With_Complex_Layouts_CVPRW_2020_paper.pdf
Search in Google Scholar Back to article
Steinhoff, P. G. (2013). Memories of New Left Protest. Contemporary Japan, 25(2), 127–165. 10.1515/cj-2013-0007
Open DOI Search in Google Scholar Back to article