English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This article describes the results of a case study that applies NeuralNetwork-based Optical Character Recognition (OCR) to scanned images of booksprinted between 1487 and 1870 by training the OCR engine OCRopus[@breuel2013high] on the RIDGES herbal text corpus [@OdebrechtEtAlSubmitted].Training specific OCR models was possible because the necessary *ground truth*is available as error-corrected diplomatic transcriptions. The OCR results havebeen evaluated for accuracy against the ground truth of unseen test sets.Character and word accuracies (percentage of correctly recognized items) forthe resulting machine-readable texts of individual documents range from 94% tomore than 99% (character level) and from 76% to 97% (word level). This includesthe earliest printed books, which were thought to be inaccessible by OCRmethods until recently. Furthermore, OCR models trained on one part of thecorpus consisting of books with different printing dates and different typesets*(mixed models)* have been tested for their predictive power on the books fromthe other part containing yet other fonts, mostly yielding character accuracieswell above 90%. It therefore seems possible to construct generalized modelstrained on a range of fonts that can be applied to a wide variety of historicalprintings still giving good results. A moderate postcorrection effort of somepages will then enable the training of individual models with even betteraccuracies. Using this method, diachronic corpora including early printings canbe constructed much faster and cheaper than by manual transcription. The OCRmethods reported here open up the possibility of transforming our printedtextual cultural heritage into electronic text by largely automatic means,which is a prerequisite for the mass conversion of scanned books.

OCR of historical printings with an application to building diachronic corpora: A case study using the RIDGES herbal corpus