3D GIS Modeling
The project presents an integrative approach to 3D modeling of complex scenes of determinate objects. The key characteristics of the proposed data model support 3D geometry and topology for extended spatial analysis, and facilitate 3D visualization. The model supports the four geometric abstractions of spatial objects, i.e. point, line, surface and body. The representation based on two constructive elements (nodes and faces) is sufficient for the derivation of a large number of 3D topological relationships. The restrictions imposed on faces (connectivity, convexity and ordering) as well as the supplementary information maintained per object (radiometric parameters and behavior) ensures the correct rendering and enables the design of realistic (photo textured), dynamic 3D worlds. The User interaction with the database residing on the Web relies on standard Internet tools — CGI scripting, VRML and HTML documents. The description of our prototype system for querying, visualizing, and communicating with a DMBS is supported by several examples. The designed model is successfully linked to developed procedures for data collection. Using a digital photogrammetric system we can collect semi-automatically geometric and semantic data, and automatically extract texture from aerial photographs as necessary for texture mapping of roof facets or texture draping over the ground surface. We consider the approach a good step forward to an operational urban 3D GIS.
Kofler, M., 1998, R-trees for the Visualization of large 3D GIS databases, PhD thesis, TU, Graz, Austria
Abdul-Rahman, A. The design and implementation of two and three-dimensional triangular irregular network (TIN) based GIS, 2000, PhD thesis, University of Glasgow, Scotland, United Kingdom, p.250.
Abdul-Rahman, A. and Pilouk, M. Spatial Data Modeling for 3D GIS. Springer 2008 Berlin Heidelberg New York. ISBN 978-3-540-74I66-4.
Brenner, C. Modeling 3D objects using weak CSG primitives. International Archives of Photogrammetric, Remote Sensing and Spatial Information Sciences ISPRS 2004, Istanbul. Vol. 35
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution 3.0 License.