Lipschitz-Killing curvatures of excursion sets for two dimensional random fields - Archive ouverte HAL Access content directly
Journal Articles Electronic Journal of Statistics Year : 2019

Lipschitz-Killing curvatures of excursion sets for two dimensional random fields

Abstract

In the present paper, we study three geometric characteristics for the excursion sets of a two dimensional stationary random field. First we show that these characteristics can be estimated without bias if the considered field satisfies a kinematic formula, this is for instance the case for fields that are given by a function of smooth Gaussian fields or for some shot noise fields. By using the proposed estimators of these geometric characteristics, we describe some inference procedures for the estimation of the parameters of the considered field. An extensive simulation study illustrates the performances of each estimator. Then we use one of the previous estimators to build a test to determine whether a given field is Gaussian or not when compared to various alternatives. The test is based on a sparse information i.e. the excursion sets for two different levels of the considered field. Finally this test is adapted and applied in a real-data case: synthesized 2D digital mammograms.
Fichier principal
Vignette du fichier
Minkovski_functional_HAL2.pdf (1.95 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01763060 , version 1 (10-04-2018)
hal-01763060 , version 2 (04-10-2018)

Identifiers

  • HAL Id : hal-01763060 , version 2

Cite

Hermine Biermé, Elena Di Bernardino, Céline Duval, Anne Estrade. Lipschitz-Killing curvatures of excursion sets for two dimensional random fields. Electronic Journal of Statistics , 2019, 13, pp.536-581. ⟨hal-01763060v2⟩
434 View
240 Download

Share

Gmail Facebook X LinkedIn More