With a semi-automated approach, a processing pipeline for geodata is presented. With the focus on segmentation for the identification of archaeological structures, 52 km² of elevation data is used. The result of this investigation should be an automatically generated city map from digital elevation data. There are several visualisation techniques for digital elevation models. Each of the techniques has its own characteristics, but also limits. In order to use these techniques correctly, the results have to compared, to know which landscape suits which technique. To make the images more understandable, visualisation from remote sensing and methods from computer science were used together to enhance the visibility of certain landscape properties.

An attempt was made to improve existing visualizations and to create meaningful images that preserve the positive properties of individual technologies. The combination of several technologies makes it possible to see different topographical features in one image. Nevertheless, there is a danger of wrong use. Important landscape features might be missed or even removed. It is therefore of great importance that the information about the raw data, scan density, methods of generating the DEMs and the visualization methods used are well documented. Thus, the assessment of artefacts can be performed better later.

More information and results will be published on the CHNT.