Urban data visualization with Voronoi diagrams

Abstract

Filtering and clustering of the data are very important aspects in data visualization. We will concentrate on these two topics and study how can we combine them to simulate a multiresolution scheme. We will focus on the properties of Voronoi Diagrams in order to avoid the need to compute any other time- or space-consuming data structure. Voronoi Diagrams capture deeply the notion of proximity between elements in an environment and allow queries to be efficiently performed. In this paper we present an application of Voronoi Diagrams and their use in visualization of georreferenced data. The input is a 2.5 data-set, and the output is a colored map where proximity to the given locations is used in order to compute the region contours. We have implemented the proposed techniques in C++. Examples of the results obtained with our application GeoVyS are given in this paper. © 2008 Springer-Verlag Berlin Heidelberg.