Enhancing text recognition of damaged documents through synergistic OCR and large language models

Asselborn, Thomas [Verfasser]; Dörpinghaus, Jens [Verfasser]; Kausar, Faraz [Verfasser]; Möller, Ralf [Verfasser]; Melzer, Sylvia [Verfasser]

Enhancing text recognition of damaged documents through synergistic OCR and large language models

bibb.id	784197
bibb.participation	BIBB-Mitarbeiter	de
dc.contributor.author	Asselborn, Thomas [Verfasser]	de
dc.contributor.author	Dörpinghaus, Jens [Verfasser]	de
dc.contributor.author	Kausar, Faraz [Verfasser]	de
dc.contributor.author	Möller, Ralf [Verfasser]	de
dc.contributor.author	Melzer, Sylvia [Verfasser]	de
dc.date.accessioned	2025-12-02T11:18:46Z
dc.date.available	2025-12-02T11:18:46Z
dc.date.issued	2024
dc.description.abstract	"Optical Character Recognition (OCR) remains a highly relevant area of research in pattern recognition. Its applications span various domains, including supporting reading for the visually impaired, interpreting Morse codes, capturing postal addresses, evaluating emails, scanning price tags and passports, and extracting text from digitised documents. As the volume of digitised data continues to grow, challenges arise in capturing the semantic structure of documents through logical structure analysis and providing data suitable for information retrieval to answer specific research questions. While classic OCR processes like Tesseract and OCRopus work well for contemporary digitised documents, there is room for improvement in text and word recognition of historical documents that are severely damaged. Large Language Models (LLMs) like GPT-4 can be effectively used for text recognition tasks, utilising their advanced natural language processing capabilities to interpret and reconstruct unclear or damaged text, offering potential for improving the overall text recognition process. However, challenges arise additionally when documents contain e.g. a mixture of single-column and double-column text, images and text, or words not known or blocked by the agents." (Authors‘ abstract; BIBB-Doku)	de
dc.description.statementofresponsibility	Thomas Asselborn, Jens Dörpinghaus, Faraz Kausar, Ralf Möller, Sylvia Melzer	de
dc.description.version	referiert	de
dc.format.extent	Seite 29-36	de
dc.format.illustration	Illustrationen	de
dc.format.illustration	Fotografien	de
dc.format.medium	Elektronische Ressource	de
dc.format.medium	Sammelbandbeitrag	de
dc.identifier.uri	DOI:10.15439/2024F7400
dc.identifier.uri	https://bibb-dspace.bibb.de/jspui/handle/BIBB/784197
dc.language.iso	en	de
dc.rdacarrier.code	cr	de
dc.rdacarrier.source	rdacontent	de
dc.rdacarrier.term	Online-Ressource	de
dc.rdacontent.code	txt	de
dc.rdacontent.source	rdacontent	de
dc.rdacontent.term	Text	de
dc.rdamedia.code	c	de
dc.rdamedia.source	rdacontent	de
dc.rdamedia.term	Computermedien	de
dc.relation.ispartof	Annals of Computer Science and Information Systems, (2024), Vol. 41: Communication Papers of the 19th Conference on Computer Science and Intelligence Systems (FedCSIS), September 8–11, 2024. Belgrade, Serbia / M. Bolanowski [Hrsg.] ; M. Ganzha [Hrsg.] ; L. Maciaszek [Hrsg.] ; M. Paprzycki [Hrsg.] ; D. Ślęzak [Hrsg.]
dc.subject.classification	G 2.2.1 Technologisierung	de
dc.subject.ddc	370	de
dc.subject.ddc	600	de
dc.subject.other	Informationstechnik	de
dc.subject.other	Texterkennung	de
dc.subject.other	Digitalisierung	de
dc.subject.other	Large Language Model	de
dc.title	Enhancing text recognition of damaged documents through synergistic OCR and large language models	de

Collections

Annals of Computer Science and Information Systems, (2024), Vol. 41: Communication Papers of the 19th Conference on Computer Science and Intelligence Systems (FedCSIS), September 8–11, 2024. Belgrade, Serbia / M. Bolanowski [Hrsg.] ; M. Ganzha [Hrsg.] ; L. Maciaszek [Hrsg.] ; M. Paprzycki [Hrsg.] ; D. Ślęzak [Hrsg.]
aktuell und lesenswert Ausgabe 3/2025- SBB und Zeitschriftenartikel
Foko 1/26 BIBB-VÖ

Enhancing text recognition of damaged documents through synergistic OCR and large language models

Files

Collections