• Digital Humanities and Cultural Heritage
• Publications close
• Data Paper close
• HAL-ENS-LYON close

In dieser Arbeit stellen wir einen OCR-Trainingsdatensatz für historische Drucke vor und zeigen, wie sich im Vergleich zu unspezifischen Modellen die Erkennungsgenauigkeit verbessert, wenn sie mithilfe dieser Daten weitertrainiert werden. Wir erörtern die Nachnutzbarkeit dieses Datensatzes anhand von zwei Experimenten, die die rechtliche Grundlage zur Veröffentlichung digitalisierter Bilddateien am Beispiel von deutschen und englischen Büchern des 19. Jahrhunderts betrachten. Wir präsentieren ein Framework, mit dem OCR-Trainingsdatensätze veröffentlicht werden können, auch wenn die Bilddateien nicht zur Wiederveröffentlichung freigegeben sind. We present an OCR ground truth data set for historical prints and show improvement of recognition results over baselines with training on this data. We reflect on reusability of the ground truth data set based on two experiments that look into the legal basis for reuse of digitized document images in the case of 19th century English and German books. We propose a framework for publishing ground truth data even when digitized document images cannot be easily redistributed. Fabrikation von Erkenntnis – Experimente in den Digital Humanities. Hg. von Manuel Burghardt, Lisa Dieckmann, Timo Steyer, Peer Trilcke, Niels Walkowski, Joëlle Weis, Ulrike Wuttke. Wolfenbüttel 2021—2022. (= Zeitschrift für digitale Geisteswissenschaften / Sonderbände, 5)

The dataset contains the spatial data needed to build the network of least cost camel paths between archaeological sites and watering places in the Egyptian Eastern Desert during the Ptolemaic and Roman periods. It includes the sites to be linked, the travel factors (topography, navigation and terrain surface), the modern traveler routes used for training and model validation, and the outputs. Data processing models, made on ArcGIS 10.5, are available. Data and analyses for model calibration and validation to assess its quality are also provided. The objective of this dataset is to be able to reproduce the network created in the Egyptian Eastern Desert and facilitate the application of the approach to other study areas. Although the input data is based on the Egyptian Eastern Desert, the method is nevertheless reproducible using the data processing models developed. Paper linked with the data: Maniere, L., Crepy, M., Redon, B. (2021) Building a Model to reconstruct the Hellenistic and Roman Road Networks of the Eastern Desert of Egypt, a Semi-Empirical Approach Based on Modern Travelers’ Itineraries. Journal of Computer Applications in Archaeology. In review.