M. BrunnhuberM. MayC. TraxlerG. Hesina,  R. W. Glatzl,  H. Kontrus (2017)

Using Different Data Sources for New Findings in Visualization of Highly Detailed Urban Data

Wo ist die Publikation erschienen?

REAL CORP 2017 -- PANTA RHEI -- A World in Constant Motion. Proceedings of 22nd International Conference on Urban Planning, Regional Development and Information Society.


Measurement of infrastructure has highly evolved in the last years. Scanning systems became more precise and many methods were found to add and improve content created for the analysis of buildings and landscapes. Therefore the pure amount of data increased significantly and new algorithms had to be found to visualize these data for further exploration. Additionally many data types and formats originate from different sources, such as Dibits hybrid scanning systems delivering laser-scanned point clouds and photogrammetric texture images. These are usually analyzed separately. Combinations of different types of data are not widely used but might lead to new findings and improved data exploration. In our work we use different data formats like meshes, unprocessed point clouds and polylines in tunnel visualization to give experts a tool to explore existing datasets in depth with a wide variety of possibilities. The diverse creation of datasets leads to new challenges for preprocessing, out-of-core rendering and efficient fusion of this varying information. Interactive analysis of different formats of data also has to have several approaches and is usually difficult to merge into one application. In this paper we describe the challenges and advantages of the combination of different data sources in tunnel visualization. Large meshes with high resolution textures are merged with dense point clouds and additional measurements. Interactive analysis can also create additional information, which has to be integrated precisely to prevent errors and misinterpretation. We present the basic algorithms used for heterogeneous data formats, how we combined them and what advantages are created by our methods. Several datasets evolve over time. This dynamic is also considered in our visualization and analysis methods to enable change detection. For tunnel monitoring this allows to investigate the entire history of the construction project and helps to make better informed decisions in the preceding construction phases or for repairs. Several methods are merged like the data they are based on enabling new ways of data exploration. In analyzing this new approach to look at heterogeneous datasets we come to the conclusion that the combination of different sources leads to a better solution than the sum of its parts.