OCR of historical printings of Latin texts

Springmann, Uwe; Najock, Dietmar; Morgenroth, Hermann; Schmid, Helmut; Gotscharek, Annette; Fink, Florian

doi:10.1145/2595188.2595205

Cited by 32 publications

(28 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [11], they apply different OCR methods to historical printings of Latin text and get the highest accuracies when using Ocropy. Some work on Blackletter fonts has been reported in [3] where models were trained on artificial training data and got high accuracies when tested on scanned books with Blackletter text.…”

Section: Related Workmentioning

confidence: 99%

Improving OCR of historical newspapers and journals published in Finland

Drobac

Kauppinen

Lindén

2019

Proceedings of the 3rd International Conference on Digital Access to Textual Cultural Heritage

View full text Add to dashboard Cite

This paper presents experiments on Optical character recognition (OCR) of historical newspapers and journals published in Finland. The corpus has two main languages: Finnish and Swedish and is written in both Blackletter and Antiqua fonts. Here we experiment with how much training data is enough to train high accuracy models, and try to train a joint model for both languages and all fonts. So far we have not been successful in getting one best model for all, but it is promising that with the mixed model we get the best results on the Finnish test set with 95 % CAR, which clearly surpasses previous results on this data set. CCS CONCEPTS • Applied computing → Optical character recognition.

show abstract

Section: Related Workmentioning

confidence: 99%

Improving OCR of historical newspapers and journals published in Finland

Drobac

Kauppinen

Lindén

2019

Proceedings of the 3rd International Conference on Digital Access to Textual Cultural Heritage

View full text Add to dashboard Cite

show abstract

“…Spelling error correction is a classical and important natural language processing (NLP) task, which, due to the large amount of unedited text available online, such as in tweets, blogs, and emails, has become even more relevant in recent times. Moreover, spelling error correction, in a broader meaning of the term, has also been of interest in the digital humanities where, for instance, large amounts of OCR (Optical character recognition) scanned text of historical or contemporary documents must be post-processed, or, even more generally, normalized (Mitankin et al, 2014;Springmann et al, 2014). In the same digital humanities context, spelling error correction may be important in correcting errors committed by scribes in reproducing historical documents (Reynolds and Wilson, 1991).…”

Section: Introductionmentioning

confidence: 99%

A Comparison of Four Character-Level String-to-String Translation Models for (OCR) Spelling Error Correction

Eger¹,

Brück²,

Mehler³

2016

The Prague Bulletin of Mathematical Linguistics

View full text Add to dashboard Cite

We consider the isolated spelling error correction problem as a specific subproblem of the more general string-to-string translation problem. In this context, we investigate four general string-to-string transformation models that have been suggested in recent years and apply them within the spelling error correction paradigm. In particular, we investigate how a simple ‘k-best decoding plus dictionary lookup’ strategy performs in this context and find that such an approach can significantly outdo baselines such as edit distance, weighted edit distance, and the noisy channel Brill and Moore model to spelling error correction. We also consider elementary combination techniques for our models such as language model weighted majority voting and center string combination. Finally, we consider real-world OCR post-correction for a dataset sampled from medieval Latin texts.

show abstract

“…Ocropy has been widely used for years. Springmann et al [36] apply different OCR methods to historical prints of Latin text and obtain high accuracy with Ocropy. Some work on Blackletter fonts has been reported in [3], where models were trained on artificial training data and achieved high accuracy on scanned books with Blackletter text.…”

Section: Related Workmentioning

confidence: 99%

Optical character recognition with neural networks and post-correction with finite state methods

Drobac

Lindén

2020

IJDAR

View full text Add to dashboard Cite

The optical character recognition (OCR) quality of the historical part of the Finnish newspaper and journal corpus is rather low for reliable search and scientific research on the OCRed data. The estimated character error rate (CER) of the corpus, achieved with commercial software, is between 8 and 13%. There have been earlier attempts to train high-quality OCR models with open-source software, like Ocropy (https://github.com/tmbdev/ocropy) and Tesseract (https://github.com/tesseract-ocr/tesseract), but so far, none of the methods have managed to successfully train a mixed model that recognizes all of the data in the corpus, which would be essential for an efficient re-OCRing of the corpus. The difficulty lies in the fact that the corpus is printed in the two main languages of Finland (Finnish and Swedish) and in two font families (Blackletter and Antiqua). In this paper, we explore the training of a variety of OCR models with deep neural networks (DNN). First, we find an optimal DNN for our data and, with additional training data, successfully train high-quality mixed-language models. Furthermore, we revisit the effect of confidence voting on the OCR results with different model combinations. Finally, we perform post-correction on the new OCR results and perform error analysis. The results show a significant boost in accuracy, resulting in 1.7% CER on the Finnish and 2.7% CER on the Swedish test set. The greatest accomplishment of the study is the successful training of one mixed language model for the entire corpus and finding a voting setup that further improves the results.

show abstract

OCR of historical printings of Latin texts

Cited by 32 publications

References 7 publications

Improving OCR of historical newspapers and journals published in Finland

Improving OCR of historical newspapers and journals published in Finland

A Comparison of Four Character-Level String-to-String Translation Models for (OCR) Spelling Error Correction

Optical character recognition with neural networks and post-correction with finite state methods

Contact Info

Product

Resources

About