OCR4all—An Open-Source Tool Providing a (Semi-)Automatic OCR Workflow for Historical Printings
Source abstract
Abstract
Optical Character Recognition (OCR) on historical printings is a challenging task mainly due to the complexity of the layout and the highly variant typography. Nevertheless, in the last few years, great progress has been made in the area of historical OCR, resulting in several powerful open-source tools for preprocessing, layout analysis and segmentation, character recognition, and post-processing. The drawback of these tools often is their limited applicability by non-technical users like humanist scholars and in particular the combined use of several tools in a workflow. In this paper, we present an open-source OCR software called OCR4all, which combines state-of-the-art OCR components and continuous model training into a comprehensive workflow. While a variety of materials can already be processed fully automatically, books with more complex layouts require manual intervention by the users. This is mostly due to the fact that the required ground truth for training stronger mixed models (for segmentation, as well as text recognition) is not available, yet, neither in the desired quantity nor quality. To deal with this issue in the short run, OCR4all offers a comfortable GUI that allows error corrections not only in the final output, but already in early stages to minimize error propagations. In the long run, this constant manual correction produces large quantities of valuable, high quality training material, which can be used to improve fully automatic approaches. Further on, extensive configuration capabilities are provided to set the degree of automation of the workflow and to make adaptations to the carefully selected default parameters for specific printings, if necessary. During experiments, the fully automated application on 19th Century novels showed that OCR4all can considerably outperform the commercial state-of-the-art tool ABBYY Finereader on moderate layouts if suitably pretrained mixed OCR models are available. Furthermore, on very complex early printed books, even users with minimal or no experience were able to capture the text with manageable effort and great quality, achieving excellent Character Error Rates (CERs) below 0.5%. The architecture of OCR4all allows the easy integration (or substitution) of newly developed tools for its main components by standardized interfaces like PageXML, thus aiming at continual higher automation for historical printings.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Handwritten Text Recognition Techniques · Digital Humanities and Scholarship · Interactive and Immersive Displays
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Artificial Intelligence and Applied Computer Science, University of Würzburg, 97074 Würzburg, Germany
Center for Information and Language Processing, LMU Munich, 80538 Munich, Germany
Institute for Modern Art History, University of Zurich, 8006 Zurich, Switzerland
Institute for Philosophy, University of Würzburg, 97074 Würzburg, Germany
References (24)
Show source-provided references
1. Doermann, D., and Tombre, K. (2014). Handbook of Document Image Processing and Recognition, Springer.10.1007/978-0-85729-859-1 ↗
2. Rydberg-Cox, J.A. (2009). Digitizing Latin incunabula: Challenges, methods, and possibilities. Dig. Hum. Q., 3.10.31826/9781463219222-009 ↗
3. OCR of historical printings with an application to building diachronic corpora: A case study using the RIDGES herbal corpus
4. Fischer, A., Wüthrich, M., Liwicki, M., Frinken, V., Bunke, H., Viehhauser, G., and Stolz, M. (2009, January 9–12). Automatic transcription of handwritten medieval documents. Proceedings of the 15th International Conference on Virtual Systems and Multimedia, 2009 (VSMM’09), Vienna, Austria.10.1109/VSMM.2009.26 ↗
5. Long short-term memory10.1162/neco.1997.9.8.1735 ↗
6. Graves, A., Fernández, S., Gomez, F., and Schmidhuber, J. (2006, January 25–29). Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks. Proceedings of the 23rd International Conference on Machine Learning, Pittsburgh, PA, USA.10.1145/1143844.1143891 ↗
7. Breuel, T.M., Ul-Hasan, A., Al-Azawi, M.A., and Shafait, F. (2013, January 25–28). High-Performance OCR for Printed English and Fraktur Using LSTM Networks. Proceedings of the 12th International Conference on Document Analysis and Recognition, Washington, DC, USA.10.1109/ICDAR.2013.140 ↗
8. Breuel, T.M. (2017, January 9–15). High Performance Text Recognition Using a Hybrid Convolutional-LSTM Implementation. Proceedings of the IEEE 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Kyoto, Japan.10.1109/ICDAR.2017.12 ↗
9. Comparison of OCR Accuracy on Early Printed Books using the Open Source Engines Calamari and OCRopus
10. Breuel, T.M. (2007, January 23–26). The hOCR microformat for OCR workflow and results. Proceedings of the IEEE Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), Curitiba, Paraná, Brazil.10.1109/ICDAR.2007.4377078 ↗
11. Pletschacher, S., and Antonacopoulos, A. (2010, January 23–26). The PAGE (page analysis and ground-truth elements) format framework. Proceedings of the IEEE 2010 20th International Conference on Pattern Recognition, Istanbul, Turkey.10.1109/ICPR.2010.72 ↗
12. Docker: lightweight linux containers for consistent development and deployment
13. Wick, C., Reul, C., and Puppe, F. (2018). Calamari—A High-Performance Tensorflow-based Deep Learning Package for Optical Character Recognition. Dig. Hum. Q., forthcoming.
14. Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., and Isard, M. (2016, January 2–4). Tensorflow: A system for large-scale machine learning. Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI 16), Savannah, GA, USA.
15. Phillips, I. (1996). User’s reference manual for the UW english/technical document image database III. UW-III English/Technical Document Image Database Manual, Intelligent Systems Laboratory.
16. Reul, C., Springmann, U., Wick, C., and Puppe, F. (2019, January 25–29). State of the Art Optical Character Recognition of 19th Century Fraktur Scripts using Open Source Engines. Proceedings of the DHd 2019 Digital Humanities: Multimedial & Multimodal, Mainz, Germany.
17. Smith, R. (2007, January 23–26). An overview of the Tesseract OCR engine. Proceedings of the IEEE Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), Curitiba, Paraná, Brazil.10.1109/ICDAR.2007.4376991 ↗
18. The OCRopus open source OCR system10.1117/12.783598 ↗
19. Afzal, M.Z., Krämer, M., Bukhari, S.S., Yousefi, M.R., Shafait, F., and Breuel, T.M. (2013, January 23). Robust binarization of stereo and monocular document images using percentile filter. Proceedings of the International Workshop on Camera-Based Document Analysis and Recognition, Washington, DC, USA.10.1007/978-3-319-05167-3_11 ↗
20. Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., and Lerer, A. (2017, January 8). Automatic Differentiation in PyTorch. Proceedings of the NIPS Autodiff Workshop, Long Beach, CA, USA.
21. Chen, K., Seuret, M., Liwicki, M., Hennebert, J., and Ingold, R. (2015, January 23–26). Page segmentation of historical document images with convolutional autoencoders. Proceedings of the IEEE 2015 13th International Conference on Document Analysis and Recognition (ICDAR), Nancy, France.10.1109/ICDAR.2015.7333914 ↗
22. Reul, C., Göttel, S., Springmann, U., Wick, C., Würzner, K.M., and Puppe, F. (2019, January 8–10). Automatic Semantic Text Tagging on Historical Lexica by Combining OCR and Typography Classification. Proceedings of the 3rd International Conference on Digital Access to Textual Cultural Heritage, Brussels, Belgium.10.1145/3322905.3322910 ↗
23. Kiessling, B. (2019, January 9–12). Kraken—An Universal Text Recognizer for the Humanities. Proceedings of the DH 2019 Digital Humanities: Complexities, Utrecht, The Netherlands.
24. Improving OCR Accuracy on Early Printed Books by combining Pretraining, Voting, and Active Learning