Share Email Print

Proceedings Paper

An extending technique of visualization of the terrain based on LoD of fractal
Author(s): Guangtang Zhu; Cheng Wang
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

To meet the special requirement in dynamic visualization of the terrain such as excavating and filling, an extending technique of the terrain visualization based on Fractal LoD (Level of Detail) is proposed. Firstly, a dynamic triangulation algorithm for the view-dependent and the real-time LoD model of the terrain is present, which is based on the GRID DEM, constructs the continuous LoD with bintree structure according to the screen error criterion, and the 'crack' is covered by building the dependent relationships of the different vertexes based on the directed acyclic graph (DAG), which aim is to render the whole terrain with the static data. Secondly, an extending algorithm based on the fractal is proposed, which can render more detail part of a special region. At beginning, the fractal dimension is calculated according the flatness of the regional surface, then 'Diamond-Square' algorithm is disposed to get the extending vertexes of the deformed region and its transitional regions according the fractal dimension and refinement degree. In order to solute the problem of the 'crack' and 'T-junction' caused by the discontinuities between the two steps, an extensional DAG and an extensional bintree structure based on the extensional vertexes are build according a extensional screen error criterion which is smaller. Lastly, the complete implement method is given.

Paper Details

Date Published: 10 November 2007
PDF: 6 pages
Proc. SPIE 6795, Second International Conference on Space Information Technology, 67957R (10 November 2007); doi: 10.1117/12.775467
Show Author Affiliations
Guangtang Zhu, Huazhong Univ. of Science and Technology (China)
Cheng Wang, Huazhong Univ. of Science and Technology (China)

Published in SPIE Proceedings Vol. 6795:
Second International Conference on Space Information Technology
Cheng Wang; Shan Zhong; Jiaolong Wei, Editor(s)

© SPIE. Terms of Use
Back to Top