Maximilian Nöth, B.A.

University of Würzburg
Centre for Philology and Digitality

Emil-Hilb-Weg 23
Campus Hubland Nord
D-97074 Würzburg

Room: 00.001

maximilian.noeth@uni-wuerzburg.de

Publications

Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning. Reul, Christian; Wick, Christoph; Nöth, Maximilian; Büttner, Andreas; Wehner, Maximilian; Springmann, Uwe. In 6th International Workshop on Historical Document Imaging and Processing. 2021.
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
In order to apply Optical Character Recognition (OCR) to historical printings of Latin script fully automatically, we report on our efforts to construct a widely-applicable polyfont recognition model yielding text with a Character Error Rate (CER) around 2% when applied out-of-the-box. Moreover, we show how this model can be further finetuned to specific classes of printings with little manual and computational effort. The mixed or polyfont model is trained on a wide variety of materials, in terms of age (from the 15th to the 19th century), typography (various types of Fraktur and Antiqua), and languages (among others, German, Latin, and French). To optimize the results we combined established techniques of OCR training like pretraining, data augmentation, and voting. In addition, we used various preprocessing methods to enrich the training data and obtain more robust models. We also implemented a two-stage approach which first trains on all available, considerably unbalanced data and then refines the output by training on a selected more balanced subset. Evaluations on 29 previously unseen books resulted in a CER of 1.73%, outperforming a widely used standard model with a CER of 2.84% by almost 40%. Training a more specialized model for some unseen Early Modern Latin books starting from our mixed model led to a CER of 1.47%, an improvement of up to 50% compared to training from scratch and up to 30% compared to training from the aforementioned standard model. Our new mixed model is made openly available to the community.

@article{reul2021mixed, abstract = {In order to apply Optical Character Recognition (OCR) to historical printings of Latin script fully automatically, we report on our efforts to construct a widely-applicable polyfont recognition model yielding text with a Character Error Rate (CER) around 2% when applied out-of-the-box. Moreover, we show how this model can be further finetuned to specific classes of printings with little manual and computational effort. The mixed or polyfont model is trained on a wide variety of materials, in terms of age (from the 15th to the 19th century), typography (various types of Fraktur and Antiqua), and languages (among others, German, Latin, and French). To optimize the results we combined established techniques of OCR training like pretraining, data augmentation, and voting. In addition, we used various preprocessing methods to enrich the training data and obtain more robust models. We also implemented a two-stage approach which first trains on all available, considerably unbalanced data and then refines the output by training on a selected more balanced subset. Evaluations on 29 previously unseen books resulted in a CER of 1.73%, outperforming a widely used standard model with a CER of 2.84% by almost 40%. Training a more specialized model for some unseen Early Modern Latin books starting from our mixed model led to a CER of 1.47%, an improvement of up to 50% compared to training from scratch and up to 30% compared to training from the aforementioned standard model. Our new mixed model is made openly available to the community.}, author = {Reul, Christian and Wick, Christoph and Nöth, Maximilian and Büttner, Andreas and Wehner, Maximilian and Springmann, Uwe}, journal = {6th International Workshop on Historical Document Imaging and Processing}, keywords = {myown}, title = {Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning.}, year = 2021 }
OCR bei Inkunabeln – Offizinspezifischer Ansatz der Universitätsbibliothek Würzburg. Kirchner, Felix; Dittrich, Marco; Beckenbauer, Phillip; Nöth, Maximilian. In ABI Technik, 36(3), pp. 178–188. 2016.
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
In the context of the Kallimachos project at the University of Würzburg, the textual base for digital edi-tions of incunabula is obtained via OCR. The corpus of texts to be worked on consists of German, French, and La-tin opera. This article shows how problems with OCR of incunabula can be tackled with already existing methods and programs. The developed method focuses on setting up a type-specific OCR training to be reused with different medieval printings from one printshop. Following this method we achieved letter accuracies of up to 95 percent and word accuracies of up to 73 percent

@article{kirchner2016inkunabeln, abstract = {In the context of the Kallimachos project at the University of Würzburg, the textual base for digital edi-tions of incunabula is obtained via OCR. The corpus of texts to be worked on consists of German, French, and La-tin opera. This article shows how problems with OCR of incunabula can be tackled with already existing methods and programs. The developed method focuses on setting up a type-specific OCR training to be reused with different medieval printings from one printshop. Following this method we achieved letter accuracies of up to 95 percent and word accuracies of up to 73 percent}, author = {Kirchner, Felix and Dittrich, Marco and Beckenbauer, Phillip and Nöth, Maximilian}, journal = {ABI Technik}, keywords = {myown}, number = 3, pages = {178-188}, title = {OCR bei Inkunabeln – Offizinspezifischer Ansatz der Universitätsbibliothek Würzburg}, volume = 36, year = 2016 }

Talks

Nöth, Maximilian. “Flexibles, skalierbares und nachhaltiges Arbeiten mit kulturellen Forschungsdaten im Semantic MediaWiki.” Culture Community Plenary, 19 Nov. 2021, Lightning Talk.

Vita

Job History

Since 11/2018: Research associate at the Centre for Philology and Digitality (University of Würzburg)
01/2018-12/2019: Research associate, project "Opera Camerarii” at the Institute of German Philology, Medieval Studies (University of Würzburg)
10/2018-12/2018: Student research assistant, project "Frühneuzeitliche Ärztebriefe des deutschsprachigen Raums (1500-1700)" at the Institute for History of Medicine (University of Würzburg)
12/2014-12/2017: Student research assistant at the University Library (University of Würzburg)

Education

2013-2017: Studies of Digital Humanities (Bachelor) at the University of Würzburg. Bachelor Thesis: "Strategonon" – Zur Digitalisierung eines frühen Kriegsspiels"

Publications

Talks

Vita

Job History

Education

Hinweis zum Datenschutz

Hinweis zum Datenschutz

Bildnachweise