OCR-D: Robust methods for layout analysis
Project object
The project aims to improve the quality and robustness of methods for the layout analysis of historical documents in OCR-D in order to ensure their practical suitability for mass digitisation. To this end, existing approaches will be optimised and expanded, and promising new procedures will be integrated. The main focus of the work lies in the further development of complementary methods for layout analysis based on artificial intelligence. The developments are accompanied by an evaluation based on scientifically established metrics. Last but not least, it is important to ensure that all methods are provided with appropriate interfaces and integrated into the overall OCR-D framework. This makes it possible, on the one hand, to flexibly combine the procedures to achieve the best possible results and, on the other hand, to guarantee their adaptability and future-proofing with regard to new developments.
Project term
2023 - 2025
Project participants
- Sächsische Landesbibliothek – Staats- und Universitätsbibliothek Dresden (SLUB Dresden)
- Zentrum für Philologie und Digitalität “Kallimachos” - Universität Würzburg (ZPD)
Third-party-funding
Deutsche Forschungsgemeinschaft (DFG)
Contact
Clemens Neudecker
Generaldirektion
Tel.: +49 30 266 434 081
clemens.neudecker@sbb.spk-berlin.de
Read more about the project
The project is part of OCR-D: DFG-funded Initiative for Optical Character Recognition Development
