Second-order analysis by variograms for curvature measures of two-phase structures

Statistical and Nonlinear Physics

DOI: 10.1140/epjb/e2005-00338-5

Cite this article as:
Arns, C., Mecke, J., Mecke, K. et al. Eur. Phys. J. B (2005) 47: 397. doi:10.1140/epjb/e2005-00338-5


Second-order characteristics are important in the description of various geometrical structures occurring in foams, porous media, complex fluids, and phase separation processes. The classical second order characteristics are pair correlation functions, which are well-known in the context of point fields and mass distributions. This paper studies systematically these and further characteristics from a unified standpoint, based on four so-called curvature measures, volume, surface area, integral of mean curvature and Euler characteristic. Their statistical estimation is straightforward only in the case of the volume measure, for which the pair correlation function is traditionally called the two-point correlation function. For the other three measures a statistical method is described which yields smoothed surrogates for pair correlation functions, namely variograms. Variograms lead to an enhanced understanding of the variability of the geometry of two-phase structures and can help in finding suitable models. The use of the statistical method is demonstrated for simulated samples related to Poisson-Voronoi tessellations, for experimental 3D images of Fontainebleau sandstone and for two samples of industrial foams.

Copyright information

© EDP Sciences/Società Italiana di Fisica/Springer-Verlag 2005

Authors and Affiliations

  1. 1.Department of Applied MathematicsResearch School of Physical Sciences and Engineering, Australian National UniversityCanberraAustralia
  2. 2.Institut für Stochastik, Friedrich-Schiller-Universität JenaJenaGermany
  3. 3.Institut für Theoretische Physik Universität Erlangen-NürnbergErlangenGermany
  4. 4.Institut für StochastikFreibergGermany