Transkribus - Handwriting Text Recognition
Transkribus is a platform to perform text-recognition (both print and handwritten), do Image/ Layout Analysis and structure recognition. It was developed for historical documents and is frequently used in archives and historical projects. For the moment it is freely accessible and has over 37,000 registered users (June 2020) from all over the world. The tool/ platform was developed as a result of the EU Project tranScriptorium (2013-2015) and READ - Recognition and Enrichment of Archival Documents (2016-2019). Since July 1st 2019 the Read Project became the READ COOP with members and shareholders. Within Transkribus several tools have been integrated, created by several participating universities and institutes. Pattern Recognition and Human Language Technology (PRHLT) from the Technical University of Valencia and the CITlab Group of the University of Rostock are very active.
What separates Transkribus – a commonly used platform for the automated recognition, transcription and searching of historical documents, from OCR‐engines is the learning curve. For example, the more transcribed pages that are added, the better language patterns are understood; resulting in Character Error Rates (CER) between 10% and 25% on previously unseen handwritten material, and less than 10% when applied to similar hands, e.g. clerical texts/paid scribes, and less than 5% when trained on an individual hand.[1]
It is expected that the Table Recognition feature is to be improved in 2020 - by the NaverLabs Europe. Another expectation is that one will need to start paying a subscription or fee from the summer onward to use the platform - to maintain the system and perform the requested tasks. (As the READ-COOP is a COOP it is not designed to make a profit).
References
- ↑ Muehlberger, Guenter; Seaward, Louise; Terras, Melissa; Ares Oliveira, Sofia; Bosch, Vicente; Bryan, Maximilian; Colutto, Sebastian; Déjean, Hervé; Diem, Markus; Fiel, Stefan; Gatos, Basilis; Greinoecker, Albert; Grüning, Tobias; Hackl, Guenter; Haukkovaara, Vili; Heyer, Gerhard; Hirvonen, Lauri; Hodel, Tobias; Jokinen, Matti; Kahle, Philip; Kallio, Mario; Kaplan, Frederic; Kleber, Florian; Labahn, Roger; Lang, Eva Maria; Laube, Sören; Leifert, Gundram; Louloudis, Georgios; McNicholl, Rory; et al. (2019). "Transforming scholarship in the archives through handwritten text recognition" (PDF). Journal of Documentation. 75 (5): 954–976. doi:10.1108/JD-07-2018-0114.
External links
- Official website
- Project tranScriptorium
- READ - Recognition and Enrichment of Archival Documents
- READ COOP
- NaverLabs Europe
This article "Transkribus - Handwriting Text Recognition" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:Transkribus - Handwriting Text Recognition. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.
