Cite PERO OCR

The development of the tool has been documented in the following two conference articles:

- [When Tesseract Does It Alone: Optical Character Recognition of Medieval Texts][8]

- [When Tesseract Brings Friends: Layout Analysis, Language Identification, and

   Super-Resolution for the Optical Character Recognition of Medieval Texts][3]

- [V Novotný: When Tesseract Does It Alone: Optical Character Recognition of

  Medieval Texts. RASLAN, 2020.][8]

- [V Novotný, K Seidlová, T Vrabcová, A Horák. When Tesseract Brings Friends:

  Layout Analysis, Language Identification, and Super-Resolution for the

  Optical Character Recognition of Medieval Texts. RASLAN, 2021.][3]

Since August 2022 (see commit 7e943d0), the tool uses PERO OCR from the Brno

University of Technology instead of Google Vision AI:

- [O Kodym, M Hradiš: Page Layout Analysis System for Unconstrained Historic

  Documents. ICDAR, 2021.][10]

- [M Kišš, K Beneš, M Hradiš: AT-ST: Self-Training Adaptation Strategy for OCR

  in Domains with Limited Transcriptions. ICDAR, 2021.][11]

- [J Kohút, M Hradiš: TS-Net: OCR Trained to Switch Between Text Transcription

  Styles. ICDAR, 2021.][12]

## Notes

 [7]: https://starfos.tacr.cz/en/project/TL03000365

 [8]: https://nlp.fi.muni.cz/raslan/2020/paper1.pdf

 [9]: https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#docker

 [10]: http://doi.org/10.1007/978-3-030-86331-9_32

 [11]: http://doi.org/10.1007/978-3-030-86337-1_31

 [12]: http://doi.org/10.1007/978-3-030-86337-1_32