Towards an Automated Healing of 3d Urban Models

Three dimensional digital city models are more and more used in different domains. These models are acquired in different ways and in very different levels of detail, due to sensor limitations or simply because of requirements of the specific scenario. The tendency towards more sophisticated models including semantic information, additional attributes and links to larger datasets can be observed in recent years. The use of semantically rich standards (e.g. CityGML) for the modelling of urban space beyond the simple representation of the geometrical aspect is one of the key hints towards a rich ‘urban information space’. This information space is going to be integrated into other applications or will be the basis for sophisticated systems in the urban domain, like simulation tools. Therefore it will be essential to control and enhance the quality of these models in order to be able to meet the needs of the aforementioned urban systems. This paper is going to introduce an approach, for validation of geometry and topology and first steps towards an automated healing. The authors recommend a test suite that provides modules for specific aspects of the model quality, which can be switched on and off so that customized tests for specific models and specific scenarios can be conducted. Reporting functionality of the tool should be divided into two parts: error report and healing report. In that way the test engineer would have information about which errors occurred, which errors were healed and by which factor the quality of the model was increased. First results of a prototype system developed at HFT Stuttgart will be presented in this paper as a starting point for further research into the field of quality management of 3D city models.