Aspekty wizualizacji rzeźby terenu w czasie rzeczywistym
Wybrane pełne teksty z tego czasopisma
Aspects of real time visualization of topographic surface
The paper presents works on creation of an optimal model of data for real time visualization of topographic surface. The model should fulfill three conditions: 1) minimization of the amount of stored data, 2) dynamic adjustment of the model to the scale of imaging, 3) guarantee of obtaining required accuracy of surface reconstruction. On the basis of literature and experiments conducted a data model based on sections was selected and an algorithm for dynamic real time visualization was elaborated. The model consists In creation of dense sections which are generalized in successive steps. Thus, the model with minimum sections arises ensuring the accuracy of reconstruction assumed in advance. Individual sections are assigned so called .priorities., which determine whether they are used in next stages of model construction or not. The higher the priority the lower distance at which the section stall be further processed. The distance is calculated from the observer to the centre of gravity of the section. The method of further creation of the model based on a network of triangles consists of the following stages: 1. Generalization is performed with the aim of decreasing the number of points, on the basis of which the network of triangles will be created. The generalization covers: m Determination of the distance between the observer and each of the sections to be analysed. The number and selection of proper sections shall depend upon the scale in which the model of topographic surface will be pictured. Based on the distance and priority sections for further analysis are selected. Designation of condensed points created a broken line of sections selected at the preceding stage. These points undergo the generalization process in order to decrease the number of triangles for the further process of creation of the topographic surface visualization. The Douglas-Peucker method was selected for generalization ensuring high level of generalization and small deformations. 2. Triangulation. The number of objects which shall be created in this step depends on the level of generalization. The increase in the level of generalization shall result in an increase of the area to be pictured and, at the same time, the accuracy of the model shall decline. In order to decrease calculations, the triangulation is conducted between neighbouring sections by the method of biggest angles. On the basis of the experiments conducted it was found that this method gives the best results providing the greatest number of triangles per unit of time. 3. Shading and illumination. It follows from the assumptions that colour shall define the height (depth). Therefore, the methods based on flat shading, which assign only one colour to each surface and bring about the lack of smooth change in colour, were not further used in the analysis. On the basis of experiments conducted, the Gouraud method was selected fulfilling the assumptions and featured with the shortest time of object shading. In the opinion of the authors, the proposed model very well fulfills the conditions presented at the beginning.
Bibliogr. 10 poz.
- 1. Foley J. D., van Dam A., Feiner S. K., Hughes J. F., Phillips R. L., 2001: Wprowadzenie do grafiki kompute- rowej. WN-T.
- 2. Marciniak A., 1998: Grafika komputerowa w języku Turbo Pascal. WN Nakom, Poznań.
- 3. Menno K., Ormeling J. F., 1998: Kartografia – wizualizacja danych przestrzennych. Wydawnictwo Naukowe PWN, Warszawa.
- 4. Pajarola R., 2002: Overview of Quadtree-based Terrain Triangulation and Visualization. Technical Report UCI-ICS-02-01, Information & Computer Science, University of California Irvine.
- 5. Petrie G., Kennie T. J. M., 1991: Terrain modeling in surveying and civil engineering. McGraw-Hill, Inc., Glasgow.
- 6. Preparata. F. P., 1985: Geometria obliczeniowa – wprowadzenie. Helion, Gliwice.
- 7. Shin-Ting W. M´arquez M. R. G., 2004: A non-self-intersection Douglas-Peucker Algorithm. Journal of the Brazilian Computer Society, Vol. 9, Fac. 3, pp.67-79, Rio de Janeiro, RJ, BR.
- 8. Stateczny A., Kamiński W ., 2003: The mathematical model of 3D fairway obtained by cross-sections and orthogonal networks RBF used for steering vessels. Proceedings of the 9th International IEEE Conference on Methods and Models in Automation and Robotics MMAR 2003, Międzyzdroje.
- 9. Zhilin L., Qing Z., Gold C., 2005: Digital terrain modeling – principles and methodology. CRC Pres.
- 10. http://grafika.iinf.polsl.gliwice.pl/doc/10-A3D.pdf