Bibliography: leaves 140-146. / Target Centring Algorithms were investigated for use in the Near-Real-Time-Photogrammetry NRTP system: PHOENICS. PHOENICS, a Photogrammetric Engineering and Industrial digital Camera System, has been developed over the past three years in the Surveying Department of UCT to provide a semi-automatic system to determine three dimensional co-ordinates of surfaces and objects using a photogrammetric method. Targets are attached to an object in order to facilitate measurement of the shape, size and orientation of the object. The centre of the target uniquely defines the target co-ordinate. Target centres (from images of the same object) are used in photogrammetric models to locate the three dimensional (3-D) coordinates of the target. The accuracy of the target 3-D location is dependent on the accuracy of the target centring algorithm. A series of sub-algorithms were employed to arrive at a single target centring algorithm. Various combinations of these sub- algorithms were compared in order to obtain the optimal target centring algorithm. Three images were used to test various aspects of the target centring algorithms: their potential accuracy was tested on an image having symmetric synthetic targets their robustness was tested on an image having targets with artificial blemishes their performance in a real (noisy) environment was tested on an image with real targets on a control frame, captured by PHOENICS. When the target centring algorithms were run on the three images, target location with an accuracy of from 1/10 of a pixel for real images, to 1/1000 of a pixel for ideal synthetic targets was obtained.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uct/oai:localhost:11427/9661 |
| Date | January 1990 |
| Creators | Rubinstein, Michael |
| Contributors | Rüther, Heinz |
| Publisher | University of Cape Town, Faculty of Engineering and the Built Environment, School of Architecture, Planning and Geomatics |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Master Thesis, Masters, MSc |
| Format | application/pdf |
Page generated in 0.0018 seconds