Theodolit-based and Knowledge-based Multi-Sensor-System
Theodolit-based and Knowledge-based Multi-Sensor-System
Disciplines
Computer Sciences (50%); Mechanical Engineering (25%); Environmental Engineering, Applied Geosciences (25%)
Keywords
-
MULTI-SENSOR-SYSTEM,
KNOWLEDGE-BASED SYSTEM,
MEASUREMENT TECHNIQUES,
DIGITAL IMAGE PROCESSING
Planners in industry and governmental offices increasingly need 3D representations of objects in the built environment. Theodolite and image-based measuremement systems, using videometric methods, can help to fulfill these standards. The CCD-camera of the theodolites is capable of capturing mosaic panorama images by camera rotation. The georeferenced images can then directly be used for deformation measurements and object reconstruction. Several theodolite measurement systems were developed, however all these industrial measurement systems can only be applied if artificial targets define the points on the objects to be monitored. With this project shall be shown that a larger set of working modes can be accomplished, if the information contained in the digital images is utilized in a more flexible way. In this case a two-step approach is straightforward: 1. Detect points of interest that serve as candidates for high accuracy object reconstruction. 2. Search corresponding points in the image of the second theodolite to gain 3D data (matching). Several working modes for interes point finding and matching will be applied. Some of these working modes are based on the assumption that there are pattern on the objectsurface to be scanned, others do not need any texture. Such a theodolite measurement system is uncommonly complicated to use and will request certain skill and experience of the operator for proper selection of the measurement mode. It thus appears that this process should be supported by a knowledge-based system component for decision making. There are two components beside the operator: a Knowledge-Based System for decision support and the Videotheodolite Processing and Control. The first phase of the project will contain an extensive analysis of the problem and the knowledge that is needed for obtaining meaningful solutions. In the next phase a suitable model for representing the knowledge will be investigated. Finally tests shall demonstrate the advantages of knowledge base measurement systems.
- Joanneum Research - 30%
- Technische Universität Wien - 40%
- Technische Universität Wien - 30%
- Gerhard Paar, Joanneum Research , associated research partner
- Thomas Eiter, Technische Universität Wien , associated research partner
Research Output
- 7 Citations
- 2 Publications
-
2012
Title Vision-Based Terrestrial Surface Monitoring DOI 10.1007/978-3-642-25495-6_10 Type Book Chapter Author Paar G Publisher Springer Nature Pages 283-348 -
2008
Title A knowledge-based videotheodolite measurement system for object representation/monitoring DOI 10.1016/j.advengsoft.2007.05.003 Type Journal Article Author Reiterer A Journal Advances in Engineering Software Pages 821-827