EUVIMEDEUVIMED entdecken
Crossref

The anisotropy of 2D or 3D Gaussian random fields through their Lipschitz–Killing curvature densities

Hermine Biermé, Agnès Desolneux

The Annals of Applied Probability · 01.06.2025 · 35 (3) · 2031-2079

DOI: 10.1214/25-aap2165 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Annals of Applied Probability, erschienen am 01.06.2025. Verfasst von Hermine Biermé und 1 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 46 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We are interested here in modeling and estimating the anisotropy of Gaussian random fields through the geometry of their excursion sets. In order to do this, we use Lipschitz–Killing curvatures of the level sets as functions of the levels and see them as generalized processes for which we are able to obtain a joint functional central limit theorem. For 2D and 3D stationary Gaussian fields we provide explicit formulas for the Lipschitz–Killing curvature densities. Then, we can deduce geometrical equivalent of second spectral moments and anisotropy ratios that allow the estimation of the anisotropy of the underlying Gaussian field.

Zitieren

APA

Biermé, H., & Desolneux, A. (2025). The anisotropy of 2D or 3D Gaussian random fields through their Lipschitz–Killing curvature densities. The Annals of Applied Probability, 35(3), 2031-2079. https://doi.org/10.1214/25-aap2165

Vancouver

Biermé H, Desolneux A. The anisotropy of 2D or 3D Gaussian random fields through their Lipschitz–Killing curvature densities. The Annals of Applied Probability. 2025;35(3):2031-2079. doi:10.1214/25-aap2165

BibTeX
@article{Bierm2025,
  author = {Biermé, Hermine and Desolneux, Agnès},
  title = {The anisotropy of 2D or 3D Gaussian random fields through their Lipschitz–Killing curvature densities},
  journal = {The Annals of Applied Probability},
  year = {2025},
  volume = {35},
  number = {3},
  pages = {2031--2079},
  doi = {10.1214/25-aap2165}
}
Kostenlos starten →