Contour line recognition from scanned topographic maps

Abstract

In this paper we present a new method for contour lines recognition starting from scanned topographical maps. This is a difficult problem due to the presence of complex textured backgrounds and information layers overlaid on the elevation lines (e.g. grid lines, rivers, roads, buildings, etc.). Our approach uses local geometric properties to recognize the contour lines but it is substantially based on the global topology of a generic topographic map (i.e. a set of non-intersecting closed lines). Starting from a scanned map the image is first segmented by a color classification process. The resulting binary image is thinned and then vectorized using a Delaunay Triangulation where the Delaunay edges are filtered using local and global rules.