OCR-D kompakt: Ergebnisse und Stand der Forschung in der Förderinitiative

Baierer, Konstantin; Boenig, Matthias; Engl, Elisabeth; Neudecker, Clemens; Altenhöner, Reinhard; Geyken, Alexander; Mangei, Johannes; Stotzka, Rainer; Dengel, Andreas; Jenckel, Martin; Gehrke, Alexander; Puppe, Frank; Weil, Stefan; Sachunsky, Robert; Schiffer, Lena K.; Janicki, Maciej; Heyer, Gerhard; Fink, Florian; Schulz, Klaus U.; Weichselbaumer, Nikolaus; Limbach, Saskia; Seuret, Mathias; Dong, Rui; Burghardt, Manuel; Christlein, Vincent; Doan, Triet Ho Anh; Doğan, Zeki; Panzer, Jörg-Holger; Schima-Voigt, Kristine; Wieder, Philipp

doi:10.1515/bfp-2020-0024

Cited by 4 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DocExtractor’s architecture relies on an encoder-decoder (namely a modified U-Net [ 43 ] with a ResNet-18 [ 44 ] encoder) for pixel-wise segmentation. We trained this “out-of-the-box” network on our data using the recommended hyper-parameters ( , acccessed on 2 October 2022) and used it to benchmark our YOLO model, as it has specifically been proposed for processing historical documents and because its architecture is commonly used in state-of-the-art OCR systems [ 45 ] to segment pages and extract text regions, outperforming Mask-RCNN [ 6 ] as shown in [ 8 ].…”

Section: Detecting Visual Elements In the Sphaera ...mentioning

confidence: 99%

CorDeep and the Sacrobosco Dataset: Detection of Visual Elements in Historical Documents

et al. 2022

View full text Add to dashboard Cite

Recent advances in object detection facilitated by deep learning have led to numerous solutions in a myriad of fields ranging from medical diagnosis to autonomous driving. However, historical research is yet to reap the benefits of such advances. This is generally due to the low number of large, coherent, and annotated datasets of historical documents, as well as the overwhelming focus on Optical Character Recognition to support the analysis of historical documents. In this paper, we highlight the importance of visual elements, in particular illustrations in historical documents, and offer a public multi-class historical visual element dataset based on the Sphaera corpus. Additionally, we train an image extraction model based on YOLO architecture and publish it through a publicly available web-service to detect and extract multi-class images from historical documents in an effort to bridge the gap between traditional and computational approaches in historical studies.

show abstract

Section: Detecting Visual Elements In the Sphaera ...mentioning

confidence: 99%

CorDeep and the Sacrobosco Dataset: Detection of Visual Elements in Historical Documents

et al. 2022

View full text Add to dashboard Cite

show abstract

“…It implements an iterative workflow that allows for rapidly training very accurate OCR models for specific publications or publication series. Lastly, OCR-D [5,6] is a workfloworiented, modular platform integrating several OCR engines into a common architecture. Unlike OCR4all, OCR-D was developed for a technical audience, such as staff working in the digitization units of cultural heritage institutions.…”

Section: Related Workmentioning

confidence: 99%

Optical Character Recognition of 19th Century Classical Commentaries: the Current State of Affairs

Romanello¹,

Najem-Meyer²,

Robertson³

2021

Preprint

View full text Add to dashboard Cite

Together with critical editions and translations, commentaries are one of the main genres of publication in literary and textual scholarship, and have a century-long tradition. Yet, the exploitation of thousands of digitized historical commentaries was hitherto hindered by the poor quality of Optical Character Recognition (OCR), especially on commentaries to Greek texts. In this paper, we evaluate the performances of two pipelines suitable for the OCR of historical classical commentaries. Our results show that Kraken + Ciaconna reaches a substantially lower character error rate (CER) than Tesseract/OCR-D on commentary sections with high density of polytonic Greek text (average CER 7% vs. 13%), while Tesseract/OCR-D is slightly more accurate than Kraken + Ciaconna on text sections written predominantly in Latin script (average CER 8.2% vs. 8.4%). As part of this paper, we also release GT4HistComment, a small dataset with OCR ground truth for 19 th classical commentaries and Pogretra, a large collection of training data and pre-trained models for a wide variety of ancient Greek typefaces.

show abstract

Putting Users in the Loop: How User Research Can Guide AI Development for a Consumer-Oriented Self-service Portal

Binder,

Diels,

Balling

et al. 2022

Culture and Computing

View full text Add to dashboard Cite

OCR-D kompakt: Ergebnisse und Stand der Forschung in der Förderinitiative

Cited by 4 publications

References 6 publications

CorDeep and the Sacrobosco Dataset: Detection of Visual Elements in Historical Documents

CorDeep and the Sacrobosco Dataset: Detection of Visual Elements in Historical Documents

Optical Character Recognition of 19th Century Classical Commentaries: the Current State of Affairs

Putting Users in the Loop: How User Research Can Guide AI Development for a Consumer-Oriented Self-service Portal

Contact Info

Product

Resources

About