Historical Document Analysis: Datasets, Analysis, and Recent Developments

The current study of the vast number of unexplored sources available in archives and libraries using classical historical methods remains beyond our human capabilities. This material record is constantly being scanned through numerous initiatives, and has led to the increase in digital historical documents. The abundance of this digital record opens the door to the application of automated analysis methods, and ultimately presenting the historian with an opportunity to explore larger corpora with the help of computers. In this course, we will look at the specific methods of Historical Document Analysis, introducing topics such as page segmentation and text extraction, page layout analysis, typeface recognition, image extraction and analysis, as well as exploring open-source tools to streamline such processes.

Literatur:

Fischer,A., Liwicki, M., and Ingold, R. 2020, Handwritten Historical Document Analysis, Recognition, and Retrieval – State of the Art and Future Trends. World Scientific.

LV-Nr.: 3131 L 110

Mo. 14-16 Uhr

Raum: H 2051 Beginn: 17.04.2023