ANALYSIS OF DYNAMIC TERRAIN PROFILING ON ANALYTICAL PHOTOGRAMMETRIC SYSTEMS (STEREOPSIS, IMAGE, SIMULATION)
Abstract
The process of dynamic observation of stereo models to measure terrain elevation was investigated with respect to sources of error. The investigation used a designed experimental approach that incorporated several factors at multiple levels. As background to the experiment the component process of stereo compilation was reviewed. A stereo compilation operation was viewed as having three major component processes, human observer, the metric stereo images and computer assisted equipment to reconstruct the stereo models and measure terrain elevation. The experiment specifically examined the following factors: (1) Base to Height ratio of stereo models (3 levels); (2) Relative direction of profiles with respect to the air base (2 levels); (3) Human observer (4 levels); (4) Direction of observing a profile (2 levels); (5) Roughness or variation of terrain height along a profile (12 levels). Among these factors, the experiment shows that the error in compiling of terrain relief by profiling is most affected by performance of the operator and the error is also sensitive to the variation in the terrain itself. Other factors are relatively minor.
Degree
Ph.D.
Subject Area
Civil engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.