Geoscience Reference
In-Depth Information
Interactive Scale-dependent multidimensional
Point Data Selection using enhanced Polarization
Transformation
Stefan Peters
Technische Universität München, Department for Cartography
München, Germany
stefan.peters@bv.tum.de
Abstract
Different fields such as Geovisualization, Web mapping or thematic and
topographic cartography all need to incorporate a most recognizable and
faithful representation of the real world by different map objects at differ-
ent scales. The objective of this work was to enhance the existing point
selection method - the Polarization Transformation - to an automatic scale-
dependent point data selection method for multidimensional point data
sets, which is implemented in an interactive (Web-) user interface. Bene-
fits of the new method are that in the resulting point selection the global as
well as the local characteristics of the spatial point distribution and of the
spatial point density are preserved; both for 2D- and 3D- point data sets.
Within an interactive user interface the user can upload a point data set,
define either the achieved output scale or the wanted number of points to
be selected. Then the determined results using the enhanced polarization
approach are shown in 2D or 3D to the user. In this work an existing 2D
point selection evaluation method for points, based on Voronoi areas, was
enhanced for 3D point selection evaluation by using Voronoi volumes.
Thus the evaluation verified the similarity of point density and distribution
before and after the point data selection.
1- Background and objectives
The power of maps lies in their ability to abstract geographic space, and
that different levels of abstraction reveal different patterns and properties
