Key techniques applied in delivering machine-readable geodata from a scanned set of late 1800s Danish historic maps

Geoffrey Brian Groom, Gregor Levin, Stig Roar Svenningsen, Mads Linnet Perner

Research output: Contribution to conferenceConference abstract for conferenceResearchpeer-review


It is increasingly often now the case that high quality digital raster scan data are available of historic maps. However, further digital processing is required to deliver machine-readable geodata (MRG), as required for GIS-based applied research. That involves application of digital image processing (DIP) methods, which, after over five decades of development, now comprise a vast toolbox of approaches and techniques, albeit all representatives of a rather concise and constant set of core functionalities. The set of techniques applied in any specific “historic map to machine readable geodata” task is just a selective subset from that toolbox. The applied set reflects upon the map material involved, the objectives of the exercise, the applied software environment, the current trends in DIP, reported relevant work, and the predilections of those undertaking the task (possibly, DIP solutions carry a personal signature of their developer?) During 2019, the authors successfully applied DIP to produce MRG from late 1800s historic topographic maps of two areas of Denmark (ca. 300 km2) for six landuse-landcover categories (forest, heath, open water, meadow, dune-sand) (Levin et al. 2020). This paper aims to distil out of the thousands of individual operations of that DIP solution’s workflow those functional forms that it applied repeatedly, examining their usages, and the reasoning of their application. The aim is to enable software-environment-independent transfer of the lessons learnt from this Danish exercise. It is also to broach inquiry as to whether the applied DIP is an additional factor of the discourse into the methodological and analytical perspectives of producing and using machine-readable data from historical maps in landscape ecological research.

Levin, G., Groom, G.B., Svenningsen, S.R. & Perner, M.L. (2020). Automated production of spatial datasets for land categories from historical maps. Method development and results for a pilot study of Danish late-1800s topographical maps. (Scientific Report No. 389) Danish Centre for Environment and Energy, Aarhus University (Denmark).
Original languageEnglish
Publication date12 Jul 2022
Number of pages1
Publication statusPublished - 12 Jul 2022
EventIALE 2022 European Landscape Ecology Congress - Online
Duration: 11 Jul 202215 Jul 2022


ConferenceIALE 2022 European Landscape Ecology Congress

Cite this