Spatial clustering analysis based on graphics
Lists
Only you see this
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.
BibTeX
@inproceedings{wang2009spatial,
author = {Zhongyuan Wang and Zongyi He},
title = {{Spatial clustering analysis based on graphics}},
booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
year = {2009},
doi = {10.1117/12.838631},
}Loading related content...