gema.pp3: Identify Clusters in a Point Pattern Using GEMA

In aproudian2/rapt: Atom Probe Tomography Analysis

Identify Clusters in a Point Pattern Using GEMA

Description

Identify Clusters in a Point Pattern Using GEMA

Usage

## S3 method for class 'pp3'
gema(X, cluster = NULL, kde.n = 20, max.clusters = 30, threshold = 0.5)

Arguments

X	The point pattern (object of class ppp or pp3) in which to identify clusters.
cluster	The marks of X that are cluster-type points. If cluster = NULL (the default) or X is unmarked, all points are assumed to be cluster-type points.
kde.n	Number of sampling grid lines for the 3D kernel density estimate
max.clusters	The maximum number of clusters to define the cluster search initialization. See Details.
threshold	The probability threshold at which to assign a point to a cluster or to the background. Defaults to 0.5, which is usually Bayes optimal.

Details

A data.frame is returned as an attribute (attr("prob")) that contains the probabilities of assignment to the background and each cluster.

Value

A marked point pattern (of the same class as X) with marks corresponding to the identified background and each identified cluster based on threshold.

References

Zelenty, J. et al., "Detecting Clusters in Atom Probe Data with Gaussian Mixture Models", Microscopy and Microanalysis, 23 (2), 269-278 (2017): https://doi.org/10.1017/S1431927617000320

See Also

Other cluster identification functions: gema.gema(), gema(), msa()

aproudian2/rapt documentation built on Dec. 15, 2022, 4:24 a.m.