The international scholarly literature on Franz Schubert and his circle of friends has been collected continuously since the founding of the archive and has been recorded in the LISA database since 1988.


Digitalisation, Recognition and Automated Clustering of Watermarks in the Music Manuscripts of Franz Schubert

Project managers: Katharina Loose-Einfalt, Günther Koliander, Andrea Lindmayr-Brandl

Collaborators: Clemens Gubsch, Paul Gulewycz, Marlene Peterlechner

The cataloguing of the music manuscripts of Franz Schubert is a primary interest of the Viennese Research Group of the New Schubert Edition (NSE). Since the beginning of this new critical edition in the 1960s, dating was a central task as it provides supplementary information on the history of the sources. This assessment was not only based on Schubert’s handwriting but also on the paper types he used. Watermarks provide evidence about their place and time of production, and consequently lead to conclusions about the use of different paper types. Thanks to continuous source documentation, the NSE can today rely on a repository of more than 1,300 handwritten watermark tracings.

As valuable as these tracings are, some were produced in unfavorable conditions and do not allow a comparative overview. A contemporary digital visualization and indexing has long been a desideratum. With modern imaging techniques, using thermography, machine learning and signal processing, more objective results can be achieved. Basic ideas here come from the field of fingerprint recognition, where a similar attempt is made to determine from a large number of data sets which of these sets correspond to the same fingerprint. The aim of the research is to enable an automatic search for similar and identical watermarks. In addition, the trained algorithm should be able to compare handtracings and thermographic images. This will not only make it possible to verify previous results of Schubert research, but also to close the gap between analogue and digitally collected source data in general.