Abstract

Maps contain valuable cartographic information, such as locations of historical places, contour lines, building footprints, and hydrography. Extracting such cartographic information from maps (ie, creating spatial layers that can be processed in a GIS) would support multiple applications and research fields. For example, there are numerous cases in which historical maps have been used to carry out research in land-cover change and biogeography [10, 14], or urban-area development [6].
Today, thousands of such maps and map series are available in raster format (ie, digital map images) in a variety of digital archives. Previous work on extracting cartographic information from raster maps typically requires intensive user intervention for training and parameter tuning, in particular, when processing historical maps of poor graphical quality [7, 13]. Consequently, most studies on analyzing and using cartographic information from historical maps are based on time-consuming manual map digitization, which introduces subjectivity because of the limited numbers of maps that can be digitized. More advanced semi-or fully automated procedures for cartographic information extraction from historical maps would allow the advancement of such studies by including historical spatial data that cover large areas, are derived from a variety of maps, and underlie reproducible procedures. In this paper, we describe a first demonstration of an interactive approach for cartographic information extraction from raster maps that have limited graphical quality and contain thematic color layers. In particular, this approach integrates an efficient and effective image-cleaning …

Date

October 14, 2025

Authors

Yao-Yi Chiang, Stefan Leyk, Craig A Knoblock

Journal

The Ninth IAPR International Workshop on Graphics Recognition

Pages

1-4