Interpolation and Analysis in Hierarchical Surface Reconstruction
Loading...
Date
1996-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Ohio State University. Division of Geodetic Science
Abstract
The primary objective of this dissertation is to develop a method for surface interpolation from a sparse set of depth data obtained by image matching. Furthermore, the method must be able to detect surface discontinuities at the same time. This approach is of paramount importance for determining surfaces of large-scale aerial images of urban areas. The proposed procedure functions within the broader framework of the hierarchical surface reconstruction approach that evolved during the past few years at the Department of Geodetic Science and Surveying. The result of this research is a new implementation of the Graduate Non-Convexity algorithm based on the multigrid techniques. The proposed method is successful in producing a realistic surface representation and a preliminary labeling of its discontinuities. The process of the Graduate Non-Convexity algorithm, as originally proposed by Blake and Zisserman [1987], depends a great deal on the optimal setting of the control parameters. In this work, a specific procedure has been developed that determines the control parameters automatically and less subjectively. Another aspect of the proposed method is the feedback of useful information to the other modules in the hierarchical surface reconstruction system, thereby facilitating the communication between the different levels of the image pyramid.