The GeoStoch project has been originated in 1999 by the Institute of Applied Information Processing and the Institute of Stochastics at Ulm University. The goal of this project is
- to develop of mathematical methods, models and algorithms in the field of stochastic geometry and spatial statistics,
- to design, develop and extend a software library for 2D and 3D image analysis and Monte Carlo simulation of spatial stochastic models,
- to apply the elaborated techniques and tools of statistical analysis to real as well as simulated data in the framework of cooperations with industrial and academic partners.
Particularly there exist cooperations with
- France Telecom R&D, Paris
- Munich Re
- Zentrum für Sonnenenergie- und Wasserstoff-Forschung (ZSW), Ulm
- Prof. Dr. P. Bäuerle, Institute of Organic Chemistry II and Advanced Materials, Ulm University
- Prof. Dr. M. Beil, Institute of Internal Medicine I, Ulm University
- Prof. Dr. M. Kazda, Institute of Systematic Botany and Ecology, Ulm University
- Prof. Dr. T. Mattfeldt, Institute of Pathology, Ulm University
- Prof. Dr. P. Walther, Electron Microscopy Facility, Ulm University
The GeoStoch library is a platform independent, Java based library for statistical image analysis and simulation of stochastic geometry models. Efficient algorithms for the realisation of random point processes, germ grain models, tessellations as well as their typical cells are implemented. Analysis tools include the computation of Minkowski functionals, significance tests for random closed sets, statistical analysis of tessellations and (marked) point patterns. Another focus is put on extrapolation methods (Kriging). A manual of several hundred pages including many examples makes it easy to use the GeoStoch library even for beginners.
Further information about the GeoStoch library can be found on the GeoStoch homepage.