Mathematical and Statistical Simulation in Problems of the Physics of Atmosphere, Ocean, and Environmental Protection
The aim is to develop and implement a high performance library which allows us to generate random points in accordance with the given discrete probability distribution on a surface or inside geometrical shapes of complex geometry in 2D and 3D spaces. The shapes can be given by a surface triangulation, a composition of geometrical primitives (such as sphere, cone, parallelepiped, cylinder, etc.) or a bitmap image. Discrete probability distribution can be given over a particular primitive and by a table. In case of bitmap images, it can be defined by gray scale tones on the image. The GeomRandom library is intended to be used for parallel construction of adaptive meshes by discrete stochastic methods, for statistical modeling problems, image processing and so on.
Note. Abstracts are published in author's edition
Mail to Webmaster |
[Home] [Conferences] |
© 1996-2008, Siberian Branch of Russian Academy of Sciences, Novosibirsk
Last update: 06-Jul-2012 (11:52:47)