EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Treating OCR Output as a Language (TOOL): A Pipeline for OCR Post-Correction

Thomas Asselborn, Magnus Bender, Ralf Möller, Sylvia Melzer

Electronics · 18.11.2026 · 15 (22) · 5378

DOI: 10.3390/electronics15225378 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 18.11.2026. Verfasst von Thomas Asselborn und 3 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deutsche Forschungsgemeinschaft.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Optical Character Recognition (OCR) enables the digitisation and analysis of large document collections, but recognition errors remain a challenge, particularly for historical documents. This paper presents TOOL (Treating OCR Output as a Language), a modular approach that formulates OCR post-correction as a sequence-to-sequence translation task. A fine-tuned MarianMT encoder–decoder model transforms noisy OCR output into corrected text without modifying the underlying recognition engine. We evaluate TOOL on a collection of historical German vocational education and training documents spanning the period from 1871 to the present day. On 139 held-out test documents, TOOL reduces mean word error rate from 0.792 for Tesseract alone to 0.600, corresponding to a relative reduction of approximately 24.3%. Improvements are also observed in Jaccard bigram similarity, BLEU-4, METEOR, and ROUGE-2, whereas mean normalised Levenshtein distance increases. Comparisons with Hunspell, Phi-4, and Qwen2.5-VL show that the larger language and vision-language models achieve higher accuracy on several metrics. However, TOOL offers a competitive quality–runtime trade-off, with an average end-to-end processing time of 30.12 s per document, compared with 96.57 s for Qwen Transcribed. The results support the feasibility of translation-based OCR post-correction as an additional processing stage in existing OCR workflows, while generalisation to other document collections, OCR engines, and languages remains to be investigated.

Förderung

  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Asselborn, T., Bender, M., Möller, R., & Melzer, S. (2026). Treating OCR Output as a Language (TOOL): A Pipeline for OCR Post-Correction. Electronics, 15(22), 5378. https://doi.org/10.3390/electronics15225378

Vancouver

Asselborn T, Bender M, Möller R, Melzer S. Treating OCR Output as a Language (TOOL): A Pipeline for OCR Post-Correction. Electronics. 2026;15(22):5378. doi:10.3390/electronics15225378

BibTeX
@article{Asselborn2026,
  author = {Asselborn, Thomas and Bender, Magnus and Möller, Ralf and Melzer, Sylvia},
  title = {Treating OCR Output as a Language (TOOL): A Pipeline for OCR Post-Correction},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {22},
  pages = {5378},
  doi = {10.3390/electronics15225378}
}
Kostenlos starten →