WorksMathematics
Testing isotropy of random fields with values in abstract spaces
Abstract
We study the problem of testing isotropy, i.e., distributional invariance under orthogonal transformations, for random fields taking values in abstract spaces. We propose a novel approach based on computing local directional characteristics, which are uniformly distributed under the assumption of isotropy. Our work is motivated by an application in material science, where we are testing the isotropy of function-valued porosity profiles of paper sheets, whose morphology is resolved by -CT imaging. While considerably more general, the presented methodology is competitive with existing isotropy tests for real-valued random fields.
Introduction
Motivation
A random field is stationary if its distribution is shift-invariant, and it is isotropic if its distribution is invariant under any orthogonal transformation (rotation or reflection) at the origin. If the random field is both, stationary and isotropic, we call it motion-invariant. For the analysis of spatial data, the assumption of motion-invariance allows for strong simplifications [10]. In particular, the covariance between two locations depends only on their Euclidean distance. Consequently, several statistical tests have been proposed in the literature to decide whether the assumption of isotropy for random fields is justified by the data. See Weller and Hoeting [48] for an overview of nonparametric tests. In recent years, further tests have been developed, including tests based on the excursion set of a given random field [1, 3], spectral domain tests [14] (Weller and Hoeting 2020), and also isotropy tests on the sphere [42].
A critical gap is that existing literature addresses primarily real-valued random fields. While multivariate extensions are scarce (see e.g. Jona-Lasinio [23]) there is currently a lack of methodology for high-dimensional or non-Euclidean data. Driven by an application in which we analyze local porosity profiles in paper-based materials represented as point-wise functional data, we aim to bridge this gap by introducing an isotropy test designed for random fields with values in abstract spaces. The analysis of local porosity in paper-based materials is of particular interest for industrial applications because it influences effective physical properties such as air permeance [52, 27]. In general, porosity profiles are frequently used descriptors for analyzing the morphology of many functional materials, such as batteries [19, 12], fuel cells [4], or cement [41].
Also in the context of materials science, -valued random fields provide another instance of non-Euclidean random fields, emerging naturally in the 3D imaging of polycrystalline materials [25, 24, 43, 40]. In this context, the values of the random field represent the crystallographic orientations of the material in space. More generally, tensor- and spinor-valued random fields provide an appropriate mathematical framework to model phenomena at the interface between materials science and continuum physics, as well as in cosmology [30].
Dataset
The methodology in this paper was developed to test for isotropy of local volume fractions in paper sheets compressed by hard-nip calendering. Local volume fractions are computed based on experimental image data. For this purpose, we consider one 3D image of the data set measured at the PETRA III storage-ring of DESY in Hamburg [18, 50] described in Neumann et al. [35]. This material has been statistically investigated to quantify interdependencies between local morphological descriptors Neumann et al., [33] and to elucidate the impact of morphology on air permeance [16].
A 3D rendering is provided in Figure 1a. The total image consists of cubic voxels with an edge length of , where each voxel is either labeled as pore or solid. Note that the solid phase is the union of cellulose fibers. The images are smoothed in horizontal direction with a 2D Gaussian kernel with standard deviation . Doing so, we obtain an image, where each voxel contains a number between 0 and 1 representing the local volume fraction of cellulose material.

Figure 1. 3D rendering of the microstructure of a compressed paper sheet, for which we test isotropy of local volume fractions (a), functional observations defined as volume fractions read in thickness direction (b), where the -direction plays the role of in (1).
As evident from the 3D rendering in Figure 1a, the fibers are concentrated in layers parallel to the surface. When assessing isotropy, we are interested in whether the distribution is invariant under plane-preserving transformations, i.e., rotations or reflections of the paper that do not involve tilting it. Hence, we interpret this 3-dimensional image as a realization of a function-valued field on a 2-dimensional domain ,
where the corresponds to the position in through-plane direction observed at some 134 equidistant points and represents the function of local volume fraction through the paper sheet at position . This interpretation is illustrated in Figure 1b.
Testing for isotropy of this function-valued field corresponds to assessing the planar isotropy of the material at microscale. For the particular example of function-valued random fields, it is also possible to apply a test developed for spatio-temporal random fields by Li et al. (2007). However, the latter test assumes stationarity in time, which is not required in the framework developed in the present work and not given in the dataset of interest.
Outline of this paper
The core idea of our work is to recast the problem of testing for isotropy as a problem of testing for uniform distribution in directions. The only structural ingredient we require from the observation space is a pairwise dissimilarity function, formalized by a semi-metric. In normed spaces, this would naturally be the norm of the difference between points.
For each sampling location, we measure the dissimilarity between all points that lie at a fixed distance and record the direction associated with the greatest dissimilarity. We will show that under isotropy, these directions are uniformly distributed. For certain choices of the semi-metric, this procedure has conceptual parallels to methods comparing directional autocovariance, e.g., Guan et al. [17].
The remainder of the paper is structured as follows: In § 2, we establish the methodological framework of the proposed hypothesis test. In § 3, we address aspects concerning practical implementation. This includes the estimation of distributional parameters for the test statistic, the choice of hyperparameters, and an outline of the algorithmic procedure. Note that an accompanying R package is also introduced here. § 4 presents an empirical finite-sample power analysis in the scalar-valued case, drawing comparisons to the methods by Guan et al. [17] and Weller and Hoeting (2020). Subsequently, § 5 analyzes the test in the context of a function-valued random field setting, which arises from assessing planar isotropy of the microstructure in paper-based materials. We apply the procedure to a synthetic data set generated from a stochastic fiber model as well as on the real data described in § 1.2. Finally, § 6 contains concluding remarks and outlines possible future directions. Theoretical justifications of our method are deferred to the Appendix.
Notation and setup
The object of interest for this paper is a strictly stationary random field taking values in some measurable space and domain , with sample size . Without loss of generality, we assume that the grid has unit spacing .
Our methodology can be extended to continuous domains or data observed on irregular spatial locations, but we refrain from developing all three formulations in parallel and leave this open for future works, as our data are sampled on a grid and our algorithms operate in the discrete setting as well.
The principal target of this work is to develop a statistical test for the hypothesis
Recall that isotropy means that the law of the random field is invariant under rotations and reflections, i.e., if and if is an element of the orthogonal group such that , we have equality of the joint laws, i.e.
Along with the required stationarity, the field is said to be motion invariant.
In order to define our test statistic we transform the process into a new field taking values on , a set of points in with Euclidean distance from , i.e. where denotes the hypersphere in with radius and center . We suppose that is an abstract space equipped with a semi-metric which is symmetric in its arguments and satisfies if and only if [51]. Now we define
which means that identifies the point in at which the field exhibits the largest dissimilarity to within the set of candidates, and identifies the lag corresponding to .
If the maximizer in (2) is not unique, we will break ties with a uniform random variable on the set of maximizers, which is independent of everything else. The tiebreakers for different are also supposed to be mutually independent.
We fix a basis subset of lags at length given by , and suppose whenever all required neighbors of are observed. Further, we denote by the set of such , where . is thus a field taking values in . Examples of the sets as well as and are illustrated in Figure 2.

Figure 2. Illustration of the sets , , and : The points for which the random field is observed () are drawn in gray and black. The points feasible for further analysis () are shown in black. Moreover, the set of neighbors used in maximizing dissimilarity are illustrated on circles. The arrows indicate lags .
From now on, we suppose is fixed, and write and instead of and .
Methodology
Formulation of the test on isotropy
The key idea behind our method is that under motion invariance, for any fixed point in the domain, has a uniform distribution on provided the following notion applies to :
Definition 2.1. A non-empty set is called transitive under its orthogonal symmetry group if there is a generating element so that for every , there exists an orthogonal transformation with and invariant under , i.e. .
This definition is closely related to the algebraic concept of orbits. Specifically, is a transitive orbit of a finite subgroup of , namely that, which leaves invariant. In the following, we write for the transpose of with .
Example 2.1. Sets that are transitive under their respective orthogonal symmetry groups include the following:
where denotes the four possible sign configurations. For a general construction with , is the set generated by :
Furthermore, all sets with are transitive under their respective orthogonal symmetry (by reflecting along ).
Example 2.2. Note that symmetry and equal distance from the origin are not sufficient. E.g., the set
is not transitive under its orthogonal symmetry group.
Some necessary conditions for to be transitive under its orthogonal symmetry group are discussed in § 8.1 of the Appendix.
The following lemma establishes uniform distribution of under isotropy given the above condition holds for . If the condition fails, then will generally not be uniformly distributed. In § 8.2 of the Appendix we give a counter-example.
Lemma 2.1. Let be a motion-invariant random field in and let be transitive under its orthogonal symmetry group. Then, for all we have that is uniformly distributed on .
We will now develop a test which assesses the new null-hypothesis given by
A more detailed discussion regarding the relationship between and is deferred to § 2.2 below. We test uniform distribution of by adapting Pearson’s -test for spatially dependent data. To this end, we transform the categorical variables into a dummy vector , whose -th component is equal to 1 if . We denote . Define
Let be some positive semi-definite estimator of the long-run covariance matrix given by
(see § 3), and suppose denotes the weighted distribution with weights given by the eigenvalues of . Define . Our test for isotropy is based on the following decision rule:
where is the -quantile.
Before proceeding, a few remarks are in order:
Remark 2.1 (Asymptotic validity). Conditions which guarantee that this is a consistent and asymptotically valid level- test will be given in the Appendix.
Remark 2.2 (Permuting labels). If we require to be affine equivariant, then the test statistic is also affine equivariant. Particularly, this implies that permuting the labels in (which is replacing by with a suitable permutation matrix ) does not change .
Remark 2.3 (Multiple sets of lags). While the above methodology only derives a decision rule for rejecting based on investigating uniform distribution along a single set , one may consider multiple such sets , . Denote their respective cardinalities by and the corresponding dummy variables by . Stacking these dummy variables to obtain and through centering by gives an analogous decision rule.
Remark 2.4 (Testing for reflection symmetry). This work focuses on testing for isotropy, i.e., distributional invariance under all orthogonal transformations. However, there are also weaker formulations, like axial, diagonal, or complete symmetry, which limit to invariance under certain reflections only, see Lu and Zimmerman (2005) for a hypothesis test of second-order symmetry characteristics. Our approach can be extended to such weaker formulations, given that is transitive under its orthogonal symmetry group limited to the respective reflectional transformations. Similarly, one may test for rotational symmetry only.
Relationship between and
We have seen that and for the latter hypothesis, we have developed a statistical test. In simplified settings it is possible to show that the implication can be reversed, that is holds. Evidently, this will depend on the choice of and . By the generality of our approach, we cannot expect universally applicable results. Instead, we illustrate the problem on the basis of an example:
Example 2.3. Let be a stationary univariate Gaussian random field on with geometric anisotropy, meaning that its covariance function has the form
where is strictly decreasing to zero and is a positive definite matrix. We set with as described in Example 2.1. For , let be the lag maximizing
A straightforward computation yields
Observing that , we conclude that unless .
The example shows that in the Gaussian case, geometric anisotropy generally implies a non-uniform distribution of even if contains only two lags. In the exceptional case where , we can adapt the test by using a different set of lags, as evident in the following example. There, we extend the above observation to multivariate fields on higher-dimensional lattices, provided all components share the same geometric anisotropy.
Example 2.4. Let be a stationary Gaussian random field on , , with values in so that its matrix-valued covariance function has the form
where , is a symmetric positive definite matrix and are strictly decreasing. Like in the previous example we choose and select the Euclidean distance for . Then,
Furthermore, there exists a selection of lag sets , each of cardinality two, so that is uniformly distributed with respect to all lag sets if and only if is a positive multiple of the identity matrix, i.e. when is isotropic.
Remark 2.5. The construction of the field in Example 2.4 is related to independent component analysis (ICA), where multivariate observations are modeled as linear mixtures of mutually independent latent sources Hyvärinen et al. [20]. Extensions of this principle to random fields have been considered in the spatial blind source separation literature, see, e.g., Nordhausen et al. [37]. Our setting corresponds to a special case, where the mixture matrix is orthogonal.
The claims of the example as well as the construction of the lag sets are further discussed in § 8.3 of the Appendix. It supports Remark 2.3: changing lags or adding lags to will make the test applicable against a bigger class of alternatives. In the example above, if we require uniform distribution of when finding maximal dissimilarity with respect to the lag sets simultaneously.
Practical implementation
Estimation of the long-run covariance
Our test statistic is based on the mean of weakly dependent data multivariate, for which Pantle et al. [38] introduce a positive semi-definite long-run covariance estimator on a continuous domain. We apply this estimator to the lattice domain case similarly to Sherman [44] but extended to higher dimensions in a way that preserves mean-square consistency, which we prove in the Appendix.
To compute the estimator, the domain is divided into sub-domains, over which the empirical means and their empirical covariance operator are subsequently computed. To this end, select a basis sub-domain (which is, e.g., rectangular) and let , be pairwise disjoint translations of that fully intersect with the domain .
Define
The positive semi-definite long-run covariance estimator is then given by
For extensions to general weight functions and non-disjoint sub-domains, we refer to Pantle et al. [38].
User-selected parameters
The semi-metric: The primary requirement for is that it must reflect that algebraic or geometric properties of the sample space and it should be capable of measuring the specific anisotropy of interest. For example, in the analysis of polycrystalline materials, quotient spaces of play an important role, where is a subgroup of . In this case, a physically meaningful semi-metric is
see Equation (6) in Karafiátová et al. [24]. Here we interpret , and as the matrix representation of the elements in . For our data example, where , the Hilbert space of square integrable functions, we will use the standard metric , where is the -norm for square integrable functions.
It is possible to combine tests for multiple semi-metrics into a single procedure, analogous to the treatment of multiple lags discussed below and in Remark 2.3.
The spatial lags: We need to choose a radius for the set of lags . Obviously, , and there must exist integers such that , in order that . Another practical limitation is the size of the domain. If the radius is large, the feasible domain becomes small.
A large radius may also cause vanishing spatial dependence between and leading to diminished statistical power. This effect can be seen in Example 2.3: if is replaced by then for , becomes uniformly distributed. On the other hand, at small radii, depending on the data situation, the anisotropy may be only barely detectable.
So in practice the choice of is a tradeoff. In general we propose using multiple smaller lags, and potentially aggregating those into a single hypothesis test (avoiding the need for any a posteriori corrections, such as, e.g., Bonferroni or Šidák corrections), see Remark 2.3. This procedure is outlined in Algorithm 2.
For the case where is real-valued, as a rule of thumb, we suggest selecting such that for , taking if the correlation is already below at . This can be determined using a direction-independent (isotropic) correlation estimator.
The basis sub-domain : The region must be chosen large enough to adequately cover the long-range dependence of the field. The specific choice of requires careful consideration, as it directly influences both the size and the power of the test. For a detailed derivation of the asymptotically optimal domain size in the two-dimensional setting, see Sherman [44].
Algorithm
In the following, we synthesize the concepts from the preceding subsections into a step-by-step computational procedure. Our algorithms are also available in form of an R package (Baumhakel et al., [5]).
Algorithm 1 outlines the testing procedure for a single set of lags and a single semi-metric. The computations in this basic algorithm are visually summarized in Figure 3, which illustrates the local computation of maximizing lags at the grid level, and global aggregation. Furthermore, Algorithm 2 details how this approach naturally extends to a combined test for multiple sets of lags and semi-metrics by appropriately stacking the test statistics. To determine the quantiles of the weighted distribution in the final step of both algorithms, we suggest Davies’ method (Davies, [11]).

Figure 3. Illustration of Algorithm 1: Computing dissimilarity-maximizing lags at every grid point (pixel) and comparing proportions to uniform distribution. The example image is a realization of an isotropic Gaussian field.
Algorithm 1 Hypothesis test for isotropy
Select distance , lags , and semi-metric
Determine , the feasible region for optimization
For each , find
Compute according to (3)
Select basis sub-domain and non-overlapping shifted versions
Compute and its eigenvalues
Determine the -quantile of a weighted distribution with weights
Compute and compare it to the quantile
Algorithm 2 Combined hypothesis test for multiple lag sets and semi-metrics
Select multiple sets of parameters
Determine the global feasible region via the intersection of all individual feasible sets
for each parameter set do
Compute by executing Steps 3–4 of Algorithm 1
end for
Stack to form the combined test statistic
Execute Steps 5–8 of Algorithm 1 for
Finite sample behavior and power analysis
While the primary motivation of our work lies in accommodating high-dimensional and functional data objects, we will now evaluate its finite-sample performance in the scalar setting and compare to specialized tests. Several nonparametric tests for spatial isotropy have been proposed for scalar-valued fields (see e.g. Weller and Hoeting [48]), and we benchmark the empirical size and power of our approach against some of these established baselines.
To this end, we replicate the simulation experiment detailed by Weller and Hoeting (2020). Specifically, we compare our proposed test (UNI, Uniform Neighborhood Isotropy testing) to the spectral domain test developed by Weller and Hoeting (2020) (SIS) and the covariance-based test by Guan et al. [17] (GSC). Note that the GSC test also requires to select a set of spatial lags with equal lengths on which it tests for equality of spatial covariance as well as a subsampling window, while SIS does neither.
As data generating process we consider a centered, stationary Gaussian random field and draw realizations on a regular grid. The spatial dependence is governed by a Matérn covariance function, given by
where is the marginal variance, is the range parameter, is a parameter related to smoothness, and is the modified Bessel function of the second kind of order . The linear transformation matrix induces geometric anisotropy and is given by
where is the angle of the principal axis of anisotropy and is the anisotropy ratio. Starting out with an isotropic field, those parameters govern in which direction () and how much () the field is stretched in order to obtain the anisotropic one.
Following the specifications of Weller and Hoeting (2020), we set the covariance parameters to , , and . The fields are generated and evaluated on different square grids to assess the effect of the sample size: up to in 5-step increments for each side length.
The null hypothesis of strict spatial isotropy corresponds to , under which the angle is unidentifiable (set to for convenience). To evaluate the empirical power, we simulate fields under combinations of anisotropy ratios and rotation angles .
We set GSC up to test for equality of autocovariance along the lags and , as well as and . We further choose subwindow dimensions for long-run covariance estimation as . Due to constraints in computation time, GSC is not computed for domains larger than . Regarding UNI, we similarly test for uniform distribution along lags , and . For estimating the long-run covariance matrix, we set the subwindow size to here. While SIS has no user-defined parameters, square grids lead to inflated size, which is why instead of computing the statistic on the full grid, we compute it on data points only.
All tests are run in R ([39]). For GSC, the implementation of the package spTest ([47]) is used. The empirical size and power results at the nominal 0.05 significance level as well as computation times for selected sample sizes are summarized in Figure 4.
Figure 4. (a) Proposed test (UNI). (b) Weller and Hoeting (2020) (SIS). (c) Guan et al. [17] (GSC). (d) Average computation time in . Empirical size and power as well as average computation time for testing spatial isotropy at the significance level. Results are based on 1000 Monte Carlo simulations for various anisotropy ratios () and angles (). Sample size is increased in steps of 5 units per direction, from a grid of size up to . GSC is only computed up to side length 30 due to computation time. The corresponding sample size up to which GSC is computed is marked with a vertical gray line in each image.
In the investigated scalar setting, UNI and SIS exhibit similar statistical power and computation times. GSC, in contrast, achieves higher statistical power, but is substantially more expensive in computation: depending on sample size, its computation time is up to two orders of magnitude larger than that of UNI and SIS. This additional computational cost is mainly due to the more involved estimation of the long-run covariance. Moreover, both GSC and UNI require the selection of a specific set of lags and a subwindow size. SIS instead considers all lags simultaneously and avoids long-run covariance estimation via subsampling. However, this computational simplicity comes at a statistical cost, as SIS exhibits inflated test size for square grids. Overall, these results indicate that the proposed UNI test is competitive with existing methods even in the scalar setting, despite it being applicable much greater generality.
Application to fiber-based materials
Before applying our UNI procedure to the dataset, we investigate its size and power empirically on synthetic fiber data, via a stochastic fiber model. With the fiber model, we mimic porous structures of a similar type as the data.
Stochastic fiber model
We use a slightly modified version of the soft-core fiber model introduced in Section II.A of Altendorf and Jeulin [2]. The solid phase is modeled by a Boolean model , see Chapter 6 in Jeulin [22], with fibers as grains. For this purpose, let be a homogeneous Poisson point process in with intensity [26]. Moreover, let be a sequence of i.i.d. copies of the random compact set , defined as follows. For a fixed , we define a random sequence , where is the origin in . Then
where denotes the closed ball with radius centered at the origin. Here, is a model parameter that describes the thickness of the fibers. The random sequence models the medial axis of a fiber with radius . In the following, we describe the distribution of this sequence. First, the main orientation is modeled by a Bingham distribution, see Section 9.4.3 in Mardia and Jupp [31], i.e. the probability density function of is given by
for each , where , positive semi-definite, is a model parameter and denotes the normalizing constant. Then, we model . The remaining are modeled by , where conditional on , the directions form a Markov chain in . For each , the conditional probability density function is
for . The parameters are controlling the influence of the main direction and the previous direction . Note that follows a von Mises-Fisher distribution conditioned on , see Section 9.3.2 in Mardia and Jupp [31]. Finally, the entire sequence is rotated around the origin, such that the main direction of the fiber, which is , coincides with . The latter is described in detail in Section II.A of Altendorf and Jeulin [2]. With this definition of the medial axis, we have specified the model of individual fibers given in (5). The solid phase, which is the union of individual fibers, is then defined by
For the Boolean model in (7), the expected volume fraction is known, i.e.
for each Borel set , see Equation (3.15) in Chiu et al. [9]. The resulting set is stationary and -dependent where , see Equation (3.18) in Chiu et al. [9] and thus fulfills weak dependence constraints required for asymptotic validity of the test, discussed in the Appendix.
For the power study, we draw realizations of the fiber model that are similar to the data set introduced in § 1.2 and represent the microstructure of a compressed paper material. Due to the compression, the fiber cross sections have elliptical shape instead, which is why we consider the transformed random closed set
for a compression factor . Note that is again stationary and has the same volume fraction as .
The simulations are performed on a voxel grid, where the voxels are cubic with an edge length of . For simplicity, we simulate from instead of with fiber length px and cross-sectional radius px. Finally, we adapt the edge lengths of the voxels accordingly to obtain . Moreover, note that our simulations model the interior part of fiber-based materials, while the image data of paper-based materials also show the boundaries of the paper, see Figure 1a. We further reduce the resolution of the discretized images by a factor of 5. Altogether, this means that is evaluated on a grid.
To determine the number of distinct balls making up the fiber, we use
In order to imitate the volume fraction observed in image data of approximately , we estimate the intensity by means of Equation (8), where the volume fraction of the left-hand side is estimated from image data and the expected volume fraction of an individual fiber is computed by means of Monte Carlo simulations.
By the spectral theorem we can decompose in (6), where is an orthogonal, and a diagonal matrix. We parametrize and define it to be the matrix representation of rotation around the -axis by angle , and . For , the resulting Bingham distribution concentrates around the equator of the sphere. Increasing , the mass concentrates on two antipodal points on the equator. When is chosen away from 0, this circular area of concentration tilts along the -axis.
We thus obtain a setting of planar isotropy for and two ways of inducing anisotropy, where in the first situation, we increase from 0 up to a maximum of , and for the second, from 0 up to . The resulting densities for these cases as well as exemplary realizations of the stochastic fiber model are shown in Figure 5.
(a) Bingham density (, ).

(b) Slices of 3-dimensional structures simulated with the stochastic fiber model (, ). Dashed lines indicate observation windows 100 × 50 (red), 200 × 100 (blue), 400 × 200 (green). Fibers run mostly horizontally, field fulfills planar isotropy.
(c) Bingham density (, ).

(d) Slices of 3-dimensional structures simulated with the stochastic fiber model (, ). Fiber directions are concentrated to preferably extend in -direction.
(e) Bingham density (, ).

(f) Slices of 3-dimensional structures simulated with the stochastic fiber model (, ). Fiber directions are mostly horizontal if they run in -direction, but have slope if they run in -direction.
Figure 5. Illustration of structures simulated with the stochastic fiber model in case of the null hypothesis (first row), and the extreme cases of the respective alternatives (second and third row). The first column illustrates the density of the Bingham distribution used to sample the main direction of each fiber. The second column shows one instance of the generated fibers, respectively, given by slicing the 3-dimensional image across the three cardinal directions. Units are given in pixels.
We investigate the power and size empirically on three sizes of observational windows: , , and , the latter corresponding to the downscaled dimensions of the dataset. The extent in -direction is fixed to 30 px. Figure 5b illustrates these window sizes relatively to the generated fiber images.
Results
To transform the image to local directions and subsequently assess uniform distribution of those directions, we maximize the norm-difference of observations at each point and its neighbors . The lags over which the maximization is performed are chosen to be sufficiently long to reliably leave the fiber when they point orthogonally to its direction in the - plane or when the fiber is not purely horizontal, while remaining short enough to avoid frequently picking up neighboring fibers. Beyond their length, we may also choose their direction. We select one set of lags that measures anisotropy close to the cardinal directions (), one that measures close to the diagonals (), and one that is in between ():
Figure 6a illustrates these sets by overlaying them on a slice of a stochastic fiber model realization. Furthermore, we select a square sub-domain of size . The empirical size under the null hypothesis remained fairly stable for window sizes varying between and .
Figure 6. (a) Sets of lags used to evaluate isotropy of stochastically generated images. Lengths are upscaled for real dataset. (b) Proportion of selections and joint 95% asymptotic confidence intervals for each lag for paper data. Angle of radial lines corresponds to angle of lag. Length of radial lines is given by selection proportions. Red circle corresponds to the uniform distribution (1/8). Ring sections outline confidence intervals for the proportions (in radial direction). Generated by R package unisono [5]. Illustration of sets of lags H (purple), H (orange), and H (yellow) and selection proportions with confidence intervals for paper data.
Figure 7 depicts empirical size of the test, as well as power under both alternatives. Clearly, increasing sample size improves power. Furthermore, the alternative of tilting the circular region of concentration by increasing is harder to detect and the test exhibits low power for small sample size.
Figure 7. (a) Alternative hypothesis: Fibers run mostly horizontally, but concentrate in a single direction. The concentration increases with . At , the field is planar isotropic. (b) Alternative hypothesis: Fibers run mostly horizontally in -direction, but have slope in -direction. At , the field is planar isotropic. Empirical power and size of UNI applied to the simulated paper material based on 1000 repetitions at significance level (marked with a horizontal line in each plot). Kinks in the lines may be further explained by instabilities in the estimation of .
We even observe just negligible power up to a certain angle . Suppose that the measurement of the paper sample is not perfectly aligned with the -plane, but instead has a certain tilt, then its interior (with respect to -direction) structure is related to that of a synthetically generated dataset with equal to that angle of tilt. Hence, the result indicates that the testing procedure exhibits no strong sensitivity with respect to such tilts.
Evaluation of isotropy on the real data
Finally, we apply the testing procedure to the dataset outlined in § 1.2. To exclude influences caused of edge effects, the image is cropped in -direction to consider only the 60 interior pixels used to tune the stochastic fiber model. To be consistent with the choice of hyperparameters in § 5.1, we multiply all previously used lag sets and subwindow dimensions by 5.
Recall that our test requires a stationary field. Stationarity in the -plane is well expected, see e.g. § 3.1 of Neumann et al., [34]. Additionally, we test this assumption layer-by-layer (in -direction) using a stationarity test for textured images by Taylor et al., [45]. The test is implemented in the R package LS2Wstat (Taylor and Nunes, [46]). The outcome confirms that there is no signs for a violation of stationarity.
After non-parallelized execution time of approximately 10 seconds on a standard laptop (eleventh generation Intel i5-1135G7 CPU), the -value output is numerically 0 (< 2.2e-16). Hence, our results indicate that there is strong evidence of anisotropy present in the dataset. This is in line with considering commercially manufactured unbleached paper sheets subjected to hard-nip calendering [35]. For commercially manufactured paper, anisotropic effects naturally arise during the production process [36].
fig:6b further illustrates the respective proportions of how often each lag appears as a maximizer. Additionally, a joint hyperrectangular 95% asymptotic confidence region for the proportions is computed based on the asymptotic Gaussian distribution of and visualized using individual confidence intervals for each proportion. These are found using the R package mvtnorm ([15]). Note that they are thus not necessarily a superset of the retention region of the proposed hypothesis test. According to Figure 6b, it appears that the violation is more in shorter lag lengths () or along the cardinal directions rather than longer lag lengths or along the diagonals. Figure 8 illustrates the process , defined through the lags on , and compares it with our stochastic fiber models under three different parameterizations. The empirical data exhibit a pattern most similar to that of the fiber model with parameters and , which corresponds to a tendency for the fibers to be oriented vertically (parallel to the -axis).
Figure 8. Top row: Maximizing directions selected from for the paper-microstructure dataset as well as exemplary realizations of the fiber model under the null hypothesis () and the two extremal alternatives ( and , respectively). Bottom row: Legend indicating how colors used in top row correspond to elements of . The direction of the radial lines corresponds to that of the lag vector. The length of radial lines is given by the selection proportions. The red line is the length under uniform proportions ().
Discussion and future work
In the present paper, we developed a hypothesis test for assessing the isotropy of abstract-valued random fields, with a particular focus on function-valued fields. We applied this methodology to evaluate the planar isotropy of microstructures in paper-based materials. Notably, the test applies to random fields in general spaces equipped with a semi-metric. Consequently, this approach provides a tool to justify simplifying assumptions regarding the spatial dependence structure for a very general class of fields.
While similar tests exist for spatio-temporal data, our setting differs fundamentally: we consider the temporal axis to be a fixed, non-expanding interval. Therefore, we do not require stationarity in time.
Furthermore, we provide graphical diagnostic tools in the form of simultaneous confidence bands. If the test rejects the null hypothesis of isotropy, these bands help identify which spatial lags are selected significantly more or less often than expected under a uniform distribution. The finite-sample performance of the test was investigated in both scalar and function-valued (specifically, high-dimensional multivariate) settings. In the scalar case, the proposed test exhibits computation times and empirical power comparable to the SIS test (Weller and Hoeting, 2020). While the GSC test [17] is generally more powerful, it requires substantially higher computation times. For large sample sizes, both our methodology and the SIS test achieve a power comparable to the GSC test.
The implementation of the test is publicly available in the R package unisono [5], which is currently being prepared for release on CRAN.
To support our methodology, we provide a theoretical framework in the appendix. In this paper, our approach is restricted to the gridded observation setting. A natural direction for future theoretical work is extending this methodology to fields observed on a continuum or on random locations. Additional methodological improvements include implementing edge corrections and developing more sophisticated procedures for handling missing data in the interior. Currently, this is only addressed implicitly by treating missing observations as part of the domain boundary. Finally, it may be beneficial to explore self-normalization techniques to bypass the estimation of the long-run covariance, see e.g., Zhang et al. [53].
Declarations
Declaration of generative AI and AI-assisted technologies in the writing process. During the preparation of this work, the authors used ChatGPT and Claude to assist with mathematical exploration, language editing, and consistency checks. The authors independently verified all mathematical claims, derivations, and proofs and reviewed and edited all AI-assisted text. The authors take full responsibility for the content of the article.
Acknowledgements. The authors thank Karin Zojer and Ingo Manke for providing 3D image data of a compressed paper sheet. The computational results have been achieved using the Austrian Scientific Computing (ASC) infrastructure.
Appendix
Necessary properties for H(0)
The following lemmas give necessary conditions for to be transitive under its orthogonal symmetry group.
Lemma 7.1 (Equal norms and finiteness). Let be transitive under its orthogonal symmetry group. Then there exists such that for all , i.e. . In particular, is finite.
Proof. Let . By transitivity there exists with and . Hence . Thus is a subset of the sphere , which is bounded and therefore contains only finitely many points of .
Lemma 7.2 (Equal spherical Voronoi measures). Let and let be transitive under its orthogonal symmetry group, with common norm (Lemma 7.1). For define the spherical Voronoi cell
and let denote the surface measure on . Then
Proof. Let . By transitivity there exists with and . Since preserves distances and permutes , it follows that . As is invariant under , . Since all have norm , the intersection , , lies in the hyperplane and is thus a -null set. Hence the cells partition up to a -null set, which proves the claim.
Lemma 7.3 (Identical distance profiles). Let be transitive under its orthogonal symmetry group. Then the multisets
do not depend on .
Proof. Let and choose with and . Then is a permutation of , and
Hence the multisets of norms coincide for and . The same holds true for the scalar products, since by Lemma 7.1, .
Properties of M0(x)
Proof of Lemma 2.1. Let and let . Due to motion invariance, we obtain
for so that is invariant under . Note that this invariance property is also valid for . We may conclude
Take any and such that and is invariant under , which exists by assumption. Then,
Example 7.1. Let be a stationary Gaussian random field on with , where and . Let be independent copies of and define
where and act componentwise. Let , which is not transitive under its orthogonal symmetry group. Then is motion invariant, but , so is not uniformly distributed on .
Proof. We first discuss motion invariance. It suffices to show that depends only on . If , then and stem from independent copies, and the covariance is zero. If , then , and the covariance is . Since squares of odd integers are and squares of even integers are , the case occurs if and only if . Hence the covariance depends only on .
is not transitive under its orthogonal symmetry group due to Lemma 7.2. We subsequently show that is not uniformly distributed. By stationarity, we may take . Assume without loss of generality that is centered. The four relevant values are
All four are standard normal and independent, except for the pair , which has correlation . Ties occur with probability zero, so holds if and only if both and lie in the random interval
The interval is independent of , and almost surely. Hence
where is jointly normal with correlation and has standard normal marginals. Note that the law of does not depend on .
For , the four values are i.i.d., so by symmetry each of the three lags is the maximizer with probability . It therefore suffices to show that is strictly increasing in for every bounded interval of positive length. By Mehler’s formula [7 Theorem 2.4.1],
where is the standard normal density and are the Hermite polynomials. All coefficients are nonnegative. Moreover, for some , since otherwise would be orthogonal to all nonconstant Hermite polynomials and thus almost everywhere constant. Hence is strictly increasing in , and for .
Relation between and
The following results support the claims in Example 2.4.
Proposition 7.1. Consider the setup described in Example 2.4. Then, for all ,
Proof. Since is orthogonal, we have for any that when is the Euclidean distance. We may therefore assume , so that the components are independent and univariate fields as in Example 2.3, with covariance functions . By stationarity it suffices to consider .
Define and as in Example 2.3, with replaced by . Then
where the pairs are independent and . Since every is strictly decreasing, all these covariances share the sign of .
Next, we write as a linear combination of random variables. To this end, denote and , which are independent, centered Gaussian variables. Then it can be easily verified that
and consequently there exist constants so that we may write
where . Moreover, it can readily be verified that .
Suppose now , so that for all . Set . Exchanging and shows . Moreover, . Hence
If , then , so and . If , then at for all we have . By continuity, this holds on an open neighborhood, which has positive probability. Hence . The case follows by exchanging and .
Corollary 7.1. Let be as in Proposition 7.1 on , , and for define
Each of these sets is transitive under its orthogonal symmetry group. If is not a positive multiple of the identity matrix, then is not uniformly distributed with respect to at least one of these lag sets. Conversely, if is a positive multiple of the identity, then is isotropic and is uniform with respect to all of them. Hence if uniformity of is required with respect to all of these lag sets simultaneously.
Proof. As in Example 2.1 and are transitive under their respective orthogonal symmetry groups. By Proposition 7.1, is uniform on if and only if . For the sets above,
If is uniform with respect to all of these sets, then all diagonal entries of coincide and all off-diagonal entries vanish, i.e. with by positive definiteness, where is the identity matrix. Conversely, if , then depends on only through , so is isotropic and is uniform by .
Asymptotic justification under the null
We assume a weak dependence structure for the field based on -mixing. To this end, suppose and denote the -algebra generated by by . Then we require
at a certain rate, where the supremum is taken over all events and and all sets , such that . Controlling dependence decay by an -mixing condition is common in isotropy testing and utilized by both Guan et al. [17] and Weller and Hoeting (2020). Furthermore, -mixing is a comparably weak assumption within classical mixing conditions, see e.g. Bradley [6]. If we assume that and is a Gaussian random field on , then implies the mixing condition , see Section 2.1.1 in Doukhan, [13]. Furthermore, -dependent fields fulfill the -mixing assumption (at any rate).
For asymptotic statements, the observation domain is denoted by and is assumed to grow with . We assume that is sufficiently regular so that for most , the full set of neighbors is available. Analogously to , we denote the set of with this property by . Formally we require the domain to grow regularly in the following sense (Nahapetian, [32], p. 154):
Assumption 1. The sets (thus also ) form a Van Hove sequence in .
Note that Assumption 1 particularly ensures that . For a detailed discussion of Van Hove sequences in and , we refer to Chapter 3.1 in Bulinski and Shashkin [8].
Theorem 7.1. Let be a motion-invariant -valued random field on such that for some . Suppose that Assumption 1 holds and that is transitive under its orthogonal symmetry group. Then,
converges in law to a zero mean Gaussian vector,
The matrix series defining is absolutely convergent.
To show Theorem 7.1, we use a central limit theorem by Nahapetian [32]. The following result is a special case of this theorem.
Lemma 7.4. Let be a real-valued centered, stationary random field on such that for some , that fulfills an -mixing condition with for some , where
Then for any sequence of sets growing in the Van Hove sense, the central limit theorem
holds, where
:::`
Proof of Theorem 7.1. The mixing condition of transfers to and thus also to the process , for any . In particular we obtain
for each . Furthermore, stationarity of also implies stationarity of and , which is an almost surely bounded random field, so the moment condition applies for arbitrary . By Lemma 7.4 we obtain that when . The multivariate CLT follows by the Cramér-Wold device.
Corollary 7.2. Suppose that is a weakly consistent estimator of and that the assumptions of Theorem 7.1 hold. Then (4) is an asymptotically valid level- test for .
Asymptotic justification under the alternative
We establish consistency for the weaker null hypothesis
The asymptotic power under fixed and local alternatives is summarized in the following two propositions. The first is a consequence of the law of large numbers.
Proposition 7.2 (Consistency against fixed alternatives). Suppose that is an anisotropic random field in that fulfills the remaining conditions of Theorem 7.1, with non-uniformly distributed in the sense that
Let be observed on according to Assumption 1. Then, the probability of rejecting tends to 1 as .
Next, we formulate a consistency result for local alternatives, i.e. settings, where anisotropy weakens with increasing sample size.
Proposition 7.3 (Power against local alternatives). Let be a sequence of stationary random fields where is observed on following Assumption 1. Let be its associated index field. Suppose that
Let be the long-run covariance of and assume as . Let further be weakly consistent estimators in the sense that , in probability. Suppose for some . Then, at significance level ,
Proof. By (9),
The bounds translate to the indicator fields and allow the triangular-array CLT of Jenish and Prucha [21] to be applied to every scalar projection. Hence, for every ,
since . Note that by Assumption 1. By the Cramér–Wold device, along every subsequence ,
Since , every subsequence admits a further subsequence converging to a limit with norm 1. Moreover, we have . Thus, by Slutsky’s theorem, converges along every such subsequence to a specific weighted non-central -distribution, which stochastically strictly dominates the identically-weighted central -distribution. This implies our claim. □
Consistency of the long-run covariance estimator
To show consistency, we require the following conditions for the basis sub-domains and shifts.
Assumption 2. For each , let be a finite lattice set forming a Van Hove sequence, with . Let , , be pairwise disjoint translates of . Moreover, assume that as .
Theorem 7.2. Let be a stationary -valued random field on that fulfills a mixing condition with for some . Let be a sequence of families of subdomains for which Assumption 2 holds. Then the estimator as described in Section 3.1 is weakly consistent.
Proof of Theorem 7.2. The estimator is positive semi-definite by the representation
for and , where indices after the semicolon denote matrix and vector entries, respectively.
To show consistency, we adapt the approach of Pantle et al. [38] (specifically Lemma 2 and Theorem 3) to a random field observed on a lattice under a weak dependence condition. We establish entry-wise mean-square consistency by showing that the estimator is asymptotically unbiased and that the variances of its entries vanish as .
Throughout the proof, let denote the centered index field. Since the estimator is invariant under a common deterministic shift of the block means, we may work with block means of . Write and , so that .
Since is a measurable transformation of , it satisfies the same mixing-rate bound. In particular, . Further, for every finite ,
First, decompose the estimator into
By stationarity,
::: By Bradley [6], the entries of are absolutely bounded by and are therefore absolutely summable. The Van Hove property and dominated convergence consequently yield .
For , the mixing covariance inequality and (10) give, entry-wise,
which goes to as .
Next, we show that the variances vanish. For , the mixing covariance inequality yields
where . By monotonicity,
Each summand on the right-hand side leaves at most choices for the unused indices. After collapsing the sum, by (10),
by Assumption 1. Similarly, for ,
Here, the first bound follows by expanding the covariance of the products, bounding the set-distance mixing coefficient by the sum of four pointwise coefficients as above, and dropping the restriction .
We have thus shown that , , and the variances of all entries of and vanish. Consequently, converges entry-wise in mean square to , and is therefore a weakly consistent estimator of .
References
- [1]Abaach, M., Biermé, H., Di Bernardino, E., and Estrade, A. (2025). Pixel isotropy test based on directional perimeters. Spatial Statistics 65, 100869. DOI: 10.1016/j.spasta.2024.100869.DOI
- [2]Altendorf, H. and Jeulin, D. (2011). Random-walk-based stochastic modeling of three-dimensional fiber systems. Physical Review E 83, 041804. DOI: 10.1103/PhysREV E.83.041804.
- [3]Azaïs, J.-M., Dalmao, F., and De Castro, Y. (2025). Gaussian random field’s anisotropy using excursion sets. arXiv: 2512.11085.arxiv.org/abs/2512.11085
- [4]Banerjee, R., Hinebaugh, J., Liu, H., Yip, R., Ge, N., and Bazylak, A. (2016). Heterogeneous porosity distributions of polymer electrolyte membrane fuel cell gas diffusion layer materials with rib-channel compression. International Journal of Hydrogen Energy 41, 14885–14896. DOI: 10.1016/J.IJHYDENE.2016.06.147.DOI
- [5]Baumhakel, J., Hörmann, S., and Neumann, M. (2026). R package: Unisono. URL: www.github.com/baumhakel/unisono.
- [6]Bradley, R. C. (2007). Introduction to Strong Mixing Conditions. Kendrick Press.
- [7]Bryc, W. (1995). The Normal Distribution: Characterizations. Springer. DOI: 10.1007/978-1-4612-2560-7.DOI
- [8]Bulinski, A. and Shashkin, A. (2007). Limit Theorems for Associated Random Fields and Related Systems. World Scientific. DOI: 10.1142/6555.
- [9]Chiu, S. N., Stoyan, D., Kendall, W. S., and Mecke, J. (2013). Stochastic Geometry and its Applications. J. Wiley & Sons. DOI: 10.1002/9781118658222.DOI
- [10]Cressie, N. A. C. (2015). Statistics for Spatial Data. Revised edition. J. Wiley & Sons. DOI: 10.1002/9781119115151.
- [11]Davies, R. B. (1980). Algorithm AS 155: The distribution of a linear combination of χ² random variables. Journal of the Royal Statistical Society C 29, 323–333. DOI: 10.2307/2346911. URL: www.jstor.org/stable/2346911.DOI
- [12]Dodell, L., Neumann, M., Osenberg, M., Hilger, A., Studer, G., Esser, B., Manke, I., and Schmidt, V. (2025). Quantifying local heterogeneities in the 3D morphology of X-PVMPT battery electrodes based on FIB-SEM measurements. ACS Materials Letters 7, 1914–1921. DOI: 10.1021/acs materialeslett.4c01552.arxiv.org/abs/2502.19085
- [13]Doukhan, P. (1994). Mixing: Properties and Examples. Springer. DOI: 10.1007/978-1-4612-2642-0. URL: link.springer.com/10.1007/978-1-4612-2642-0.
- [14]Eckle, T., van Delft, A., and Dette, H. (2026). Balancing the edge effect and dimension of spectral spatial statistics under irregular sampling with applications to isotropy testing. Bernoulli 32, 442–466. DOI: 10.3150/25-bej1865.DOI
- [15]Genz, A. and Bretz, F. (2009). Computation of Multivariate Normal and t Probabilities. Springer.DOI
- [16]Gräfensteiner, P., Rodriguez, A., Leitl, P., Baikova, E., Fuchs, M., Machado-Charry, E., Hirn, U., Hilger, A., Manke, I., Schennach, R., Neumann, M., Schmidt, V., and Zojer, K. (2026). Predicting air flow in calendered paper sheets from micro-CT data: Combining physics with morphology. Transport in Porous Media 153, 15. DOI: 10.1007/s11242-025-02250-9.DOI
- [17]Guan, Y., Sherman, M., and Calvin, J. A. (2004). A nonparametric test for spatial isotropy using subsampling. Journal of the American Statistical Association 99, 810–821. DOI: 10.1198/016214504000001150.DOI
- [18]Haibel, A., Beckmann, F., Dose, T., Herzen, J., Ogurreck, M., Müller, M., and Schreyer, A. (2010). Latest developments in microtomography and nanotomography at PETRA III. Powder Diffraction 25, 161–164. DOI: 10.1154/1.3428364.DOI
- [19]Hosseinzadeh, E., Marco, J., and Jennings, P. (2018). The impact of multi-layered porosity distribution on the performance of a lithium ion battery. Applied Mathematical Modelling 61, 107–123. DOI: 10.1016/J.APM.2018.04.001.DOI
- [20]Hyvärinen, A., Karhunen, J., and Oja, E. (2001). Independent Component Analysis. J. Wiley & Sons. DOI: 10.1002/0471221317.DOI
- [21]Jenish, N. and Prucha, I. R. (2009). Central limit theorems and uniform laws of large numbers for arrays of random fields. Journal of Econometrics 150, 86–98. DOI: 10.1016/j.jeconom.2009.02.009.DOI
- [22]Jeulin, D. (2021). Morphological Models of Random Structures. Springer. DOI: 10.1007/978-3-030-75452-5.DOI
- [23]Jona-Lasinio, G. (2001). Modeling and exploring multivariate spatial variation: A test procedure for isotropy of multivariate spatial data. Journal of Multivariate Analysis 77, 295–317. DOI: 10.1006/jmva.2000.1938.DOI
- [24]Karafíátová, I., Møller, J., Pawlas, Z., Staněk, J., Seitl, F., and Beneš, V. (2023). Fitting the grain orientation distribution of a polycrystalline material conditioned on a Laguerre tessellation. Spatial Statistics 55, 100747. DOI: 10.1016/j.spasta.2023.100747.DOI
- [25]Karafíátová, I., Pawlas, Z., and Heller, L. (2022). Statistical assessment of stress redistribution in loaded polycrystals. Image Analysis & Stereology 41. DOI: 10.5566/ias.2642.DOI
- [26]Last, G. and Penrose, M. (2018). Lectures on the Poisson Process. Cambridge University Press. DOI: 10.1017/9781316104477.DOI
- [27]Leitl, P., Machado-Charry, E., Baikova, E., Neumann, M., Hirn, U., Schmidt, V., and Zojer, K. (2023). Joint distributions of local pore space metrics quantitatively explain simulated air flow variations in paper. Transport in Porous Media 148, 627–648. DOI: 10.1007/s11242-023-01964-y.DOI
- [30]Malyarenko, A. and Ostoja-Starzewski, M. (2023). Tensor- and spinor-valued random fields with applications to continuum physics and cosmology. Probability Surveys 20, 1–86. DOI: 10.1214/22-ps12.DOI
- [31]Mardia, K. V. and Jupp, P. E. (2000). Directional Statistics. J. Wiley & Sons. DOI: 10.1002/9780470316979.DOI
- [32]Nahapetian, B. (1991). Limit Theorems and Some Applications in Statistical Physics. Vieweg + Teubner Verlag. DOI: 10.1007/978-3-322-93432-1. URL: link.springer.com/10.1007/978-3-322-93432-1.DOI
- [33]Neumann, M., Gräfensteiner, P., Machado-Charry, E., Hirn, U., Hilger, A., Manke, I., Schennach, R., Schmidt, V., and Zojer, K. (2024). R-vine copulas for data-driven quantification of descriptor relationships in porous materials. Advanced Theory and Simulations 2024, 2301261. DOI: 10.1002/adts.202301261.DOI
- [34]Neumann, M., Machado-Charry, E., Zojer, K., and Schmidt, V. (2021a). On variability and interdependence of local porosity and local tortuosity in porous materials: a case study for sack paper. Methodology and Computing in Applied Probability 23, 613–627. DOI: 10.1007/s11009-019-09761-1.DOI
- [35]Neumann, M., Machado-Charry, E., Baikova, E., Hilger, A., Hirn, U., Schennach, R., Manke, I., Schmidt, V., and Zojer, K. (2021b). Capturing centimeter-scale local variations in paper pore space via μ-CT: A benchmark study using calendered paper. Microscopy and Microanalysis 27, 1305–1315. DOI: 10.1017/S1431927621012563.DOI
- [36]Niskanen, K. J. (2018). Distribution of Fibre Orientations in Paper. In: Fundamentals of Papermaking. Ed. by C. F. Baker and V. Punton. FRC, 275–308. DOI: 10.15376/frc.1989.1.275.
- [37]Nordhausen, K., Oja, H., Filzmoser, P., and Reimann, C. (2015). Blind source separation for spatial compositional data. Mathematical Geosciences 47, 753–770. DOI: 10.1007/s11004-014-9559-5.DOI
- [38]Pantle, U., Schmidt, V., and Spodarev, E. (2010). On the estimation of integrated covariance functions of stationary random fields. Scandinavian Journal of Statistics 37, 47–66. DOI: 10.1111/j.1467-9469.2009.00663.x.DOI
- [39]R Core Team (2023). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. URL: www.R-project.org/.
- [40]Rieder, P., Neumann, M., Fernandes, L. M., Mulard, A., Proudhon, H., Willot, F., and Schmidt, V. (2024). Stochastic 3D microstructure modeling of twinned polycrystals for investigating the mechanical behavior of γ-TiAl intermetallics. Computational Materials Science 238, 112922. DOI: 10.1016/j.commatsci.2024.112922.DOI
- [41]Rimmelé, G., Barlet-Gouédard, V., Porcherie, O., Goffé, B., and Brunet, F. (2008). Heterogeneous porosity distribution in Portland cement exposed to CO₂-rich fluids. Cement and Concrete Research, 1038–1048. DOI: 10.1016/J.CEMCONRES.2008.03.022.DOI
- [42]Sahoo, I., Guinness, J., and Reich, B. J. (2019). A test for isotropy on a sphere using spherical harmonic functions. Statistica Sinica 29, 1253–1276. DOI: 10.5705/ss.202017.0475.DOI
- [43]Sheng, N., Khazaie, S., Chevreuil, M., and Fréour, S. (2023). Statistical properties of effective elastic moduli of random cubic polycrystals. Mechanics & Industry 24, 33. DOI: 10.1051/meca/2023030.DOI
- [44]Sherman, M. (1996). Variance estimation for statistics computed from spatial lattice data. Journal of the Royal Statistical Society Series B 58, 509–523. DOI: 10.1111/J.2517-6161.1996.TB02097.X.DOI
- [45]Taylor, S., Eckley, I., and Nunes, M. (2014). A test of stationarity for textured images. Technometrics 56, 291–301. DOI: 10.1080/00401706.2013.823890.DOI
- [46]Taylor, S. and Nunes, M. (2023). R package: LS2Wstat: A multiscale test of spatial stationarity for LS2W processes. DOI: 10.32614/CRAN.package.LS2Wstat. URL: https://CRAN.R-project.org/package=LS2Wstat.
- [47]Weller, Z. D. (2018). SpTest: An R package implementing nonparametric tests of isotropy. Journal of Statistical Software 83, 1–24. DOI: 10.18637/jss.v083.i04.DOI
- [48]Weller, Z. D. and Hoeting, J. A. (2016). A review of nonparametric hypothesis tests of isotropy properties in spatial data. Statistical Science 31, 305–324. DOI: 10.1214/16-sts547. URL: www.jstor.org/stable/24780855.DOI
- [50]Wilde, F., Ogurreck, M., Greving, I., Hammel, J. U., Beckmann, F., Hipp, A., Lottermoser, L., Khokhriakov, I., Lytaev, P., Dose, T., Burmester, H., Müller, M., and Schreyer, A. (2016). Micro-CT at the imaging beamline P05 at PETRA III. In: AIP Conference Proceedings. Vol. 1741. AIP Publishing LLC, 030035. DOI: 10.1063/1.4952858.DOI
- [51]Wilson, W. A. (1931). On semi-metric spaces. American Journal of Mathematics 53, 361–373. DOI: 10.2307/2370790.DOI
- [52]Woudberg, S., Theron, F., Lys, E., and Le Coq, L. (2018). Investigating the influence of local porosity variations and anisotropy effects on the permeability of fibrous media for air filtration. Chemical Engineering Science 180, 70–78. DOI: 10.1016/J.CES.2018.01.035.DOI
- [53]Zhang, X., Li, B., and Shao, X. (2014). Self-normalization for spatial data. Scandinavian Journal of Statistics 41, 311–324. DOI: 10.1111/sjos.12028.DOI






