Generalisation of heterogeneous topographic data to derive a German Alpine Club Map

OSM and Terrain of the region with the Ushba Mountain.

Objective:

Research tasks particularly concern the model and cartographic generalization of the heterogeneous source data. It is necessary to adapt the classification of roads, paths and water bodies in accordance with the convention of the Alpine Club cards. A workflow is to be developed for the semi-automatic generalization of linear objects, in particular selection, simplification and displacement. A challenge here is the linking of individual line segments, which are modelled in OpenStreetMap only as relations. Neighboring objects and the shape of the terrain must be taken into account during the cartographic generalization. Furthermore, identified small data gaps and errors in OpenStreetMap should be fixed using aerial and satellite images. Finally, the data is used as the basis for the field work, which serves to evaluate the intermediate mapping product, to which the candidate is cordially invited.
Workflows which have to be developed for generalisation of roads, hiking trails and water network. https://desktop.arcgis.com/en/arcmap/latest/tools/cartography-toolbox/conflict-resolution-and-generalization-with-geoprocessing.htm

Description:

We are planning to produce a map of the Ushba Mountain in Georgian at the scale of 1:33’000. It is one of the most notable peaks of the Caucasus Mountains next to the border of Russia. The task of the master's thesis is to process and generalize the topographic data for the chosen scale. The main focus here is on the cartographic representation of the linear features, such as the roads and hiking trails and the water network. OpenStreetMap is the source of the data which should be harmonized with the digital terrain model.

Staff
Dirk Burghardt
Mathias Gröbe

Implementation of a generalization workflow in in ArcGIS using the ArcGIS Model builder

References:

Domain(s):

Study Program(s):

  • MSc. Cartography (EXCLUSIVELY externally advertised)