This Article 
 Bibliographic References 
 Add to: 
On the Interpolation of Data with Normally Distributed Uncertainty for Visualization
Dec. 2012 (vol. 18 no. 12)
pp. 2305-2314
Steven Schlegel, University of Leipzig
Nico Korn, Helmholtz-Centre Dresden-Rossendorf
Gerik Scheuermann, University of Leipzig
In many fields of science or engineering, we are confronted with uncertain data. For that reason, the visualization of uncertainty received a lot of attention, especially in recent years. In the majority of cases, Gaussian distributions are used to describe uncertain behavior, because they are able to model many phenomena encountered in science. Therefore, in most applications uncertain data is (or is assumed to be) Gaussian distributed. If such uncertain data is given on fixed positions, the question of interpolation arises for many visualization approaches. In this paper, we analyze the effects of the usual linear interpolation schemes for visualization of Gaussian distributed data. In addition, we demonstrate that methods known in geostatistics and machine learning have favorable properties for visualization purposes in this case.
Index Terms:
Interpolation,Gaussian processes,Data visualization,Uncertainty,Distributed databases,Random variables,Data models,interpolation,Gaussian process,uncertainty
Steven Schlegel, Nico Korn, Gerik Scheuermann, "On the Interpolation of Data with Normally Distributed Uncertainty for Visualization," IEEE Transactions on Visualization and Computer Graphics, vol. 18, no. 12, pp. 2305-2314, Dec. 2012, doi:10.1109/TVCG.2012.249
Usage of this product signifies your acceptance of the Terms of Use.