Construction of a web-based database for the objective analysis of scribal hands of the medieval monastery library of Fulda in cooperation with the Institut Bibliotheka Fuldensis
Since the 1980s, the Institut Bibliotheca Fuldensis has been studying the library and scriptorium of Fulda Abbey in the Middle Ages. Fulda possessed one of the most important libraries of the Carolingian period, which was, however, first carried off and then largely destroyed during the 30 Years' War in 1632. The collection and study of its remains as well as the products (manuscripts) of the Fulda Scriptorium, which are today scattered from Europe to the USA, are of utmost importance in terms of cultural history.
The aim of this project is the creation and programming of a web-based database which will make it possible to objectively and unambiguously assign manuscripts to the Fulda Monastery Library and, in addition, to identify individual scribal hands through comparative statistics and to place them in a chronological framework.
The database is intended to offer the Institut Bibliotheca Fuldensis , but also palaeographers worldwide, the opportunity to work on the reconstruction of the medieval library of Fulda Abbey.
By establishing a web-based database, it will be possible for palaeographers worldwide to use and expand this database. This will open up new, internationally networked project opportunities.
The interdisciplinary approach is a key factor in the success of this project. This database will be available for manuscripts from the holdings of the University and State Library as well as for any other expert in palaeographic research.
The systematic construction of a data collection of scribal hands can be automated within certain limits. Appropriate image-processing approaches can be used to segment the digitised material. This makes it possible to recognise the positions of lines, words and, to a more limited extent, letters in the image material.
A complete handwriting recognition in the sense of an OCR (Optical Character Recognition) is rather not possible on the basis of the small data volume. However, the recognition of relevant optical properties of individual letters, the so-called feature detection, can be implemented very well. This makes it possible to assign reproducible optical properties to individual letters, which can then be evaluated palaeographically.
Thus, the interactive, web-based data collection can be used as a system for the targeted and efficient search of comparison points in the image data. Thus, the planned approach represents a substantial support of the palaeographic work process.
Projektleitung
Prof. Dr habil. Marc Birringer
Applied biochemistry for nutrition and the environment
Projektleitung
Prof. Dr. Jan-Torsten Milde Programme Director Bachelor Digital Media
Media IT and Web Technologies