A Technique for Skeletonizing a Scalar Field Using a Critical Point Graph: Application to a Weather Simulation

We present a visualization technique which a large scale detects the structure at all scales removing from the user the responsibility of extracting information implicit in the data, and presenting the structure explicitly for analysis and retrieval. In typical scientific applications, data is represented at the nodes of a mesh of elements and interpolated linearly across the interior of the elements. A critical point exists in the gradient field of the Scalar filed. It is possible at few cost to express the Scalar field using the graph which connected the critical point. In past, we proposed a technique for calculating isosurfaces efficiently using an extrema skeleton, which consists of elements and connects all extrema points [3]. In this algorithm, extrema points in a scalar field are first extracted. A graph is then generated in which the extrema points are taken as nodes. And our work can be regarded as an extension of our original technique, because our new skeleton connects saddle points in addition to extrema points. To confirm the effectiveness of our technique, we extract features skeletons of CFD results for forecasting weather and determine linear correlation between multiple scalar variables such as wind velocity, temperature, and humidity.