3D Spatial Data
Recent years have witnessed significant growth in the usage of 3D data within the geospatial industry, especially when used to provide immersive visualisation of urban data. 1Spatial believes it is inevitable that in the near future 3D data will begin to be used to perform spatial analysis and, as such, there will be an increasing demand for improvements to data quality as is now fundamental in 2D geospatial data. To this end 1Spatial is running several research projects to help learn from the lessons of the past 20 years and to identify the spatial analysis and quality tools that will be essential to support such a change of use.
3D topological model
The advantages of having a topological model underpinning the feature data are well established in the 2D world and include adding intelligence to the data and ensuring consistency of the relationship between features. 1Spatial is currently performing research driven development of such a topology model to underpin 3D geospatial data.
In 2007 Oracle added support of solid features to the Oracle Spatial™ product, which allowed the storage of 3D solid geospatial data features. 1Spatial’s aim is to take this further and develop an engine that creates and maintains a dynamic, ISO19107-compliant topological structure based on this Oracle 3D feature data. As those feature data are modified, the topological structure is maintained, removing minor topological errors (by snapping) as well as identifying more major issues (e.g. overlaps and slivers), enabling them to be addressed. This topology model allows the development of very fast topological operators (for example touch, overlaps, contains) that, as well as supplementing the standard Oracle operator set, allow the development of a series of tools for geospatial data quality and analysis on 3D data.
To demonstrate the effect of topology, part of the demonstration scenario will include the generation of network data from the topology and show how this can be dynamically updated from the modified topology
3D spatial data quality
Most of the quality checks required on geospatial data involve testing how features interact with each other and the majority of those are topological in nature, for example do two features of the same class overlap? Are all items of one class contained in items of another? Normally such checks are performed on-the-fly using geospatial and geometric calculations that, for very large datasets, can be very time-intensive.
If a topology database underpins the feature data, these topological queries become encoded in the database and are therefore imwww-ext-testtely accessible to the data quality tools. This brings topological quality checks into the realms of real-time. 1Spatial is currently investigating and developing a series of data quality services to perform the following checks on 3D data:
- Detect overlapping solids/polygons
- Detect voids between solids
- Ensure containment (for example Check all rooms are within a building)
- Face-solid match (Ensure all polygons lie wholly on the face of a solid)
- Detect overshoots (of planes)
- Enforce shared topology (Check that all buildings touch the ground)
The Face-solid match case needs to be noted as it is fundamental in models that contain solid models and separate polygons representing textures. Such models have an inherent maintainability problem in keeping the solid and polygons in step. This service allows the checking of such data.
Customer support:
1Spatial news RSS feed 
