Share Email Print
cover

Proceedings Paper

Spatial clustering analysis based on graphics
Author(s): Zhongyuan Wang; Zongyi He
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Most of the traditional clustering methods are based on spatial database, which reach the clustering results by operating and computing the spatial coordinates and its attribute values in database. So it can not consider such factors of the clustering objects as the morphological characters, topological relation etc in the process and results of clustering. From the point of view of spatial geometry, this paper explored to build the basis of clustering analysis on graphics, divided spatial clustering into eight types, put forward some kinds of methods of model simplification and dimension reduction based on spatial projection and mapping, and at last demonstrated the graphics-based spatial clustering method by introduced a fast spatial clustering method based on Graph Theory and Delaunay triangulation, which achieved good results judged by the experiment data.

Paper Details

Date Published: 16 October 2009
PDF: 9 pages
Proc. SPIE 7492, International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining, 74924Q (16 October 2009); doi: 10.1117/12.838631
Show Author Affiliations
Zhongyuan Wang, China Univ. of Mining and Technology (China)
Wuhan Univ. (China)
Zongyi He, Wuhan Univ. (China)


Published in SPIE Proceedings Vol. 7492:
International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining
Yaolin Liu; Xinming Tang, Editor(s)

© SPIE. Terms of Use
Back to Top