lacunaritycovariance (version 1.1-7)

secondorderprops: Estimate Second-Order Properties of a RACS

Description

Estimates many second order properties of RACS, gliding box lacunarity, covariance, centred covariance, and pair-correlation. This can be faster than computing estimates of multiple second order properties separately as Fourier transforms of the binary map are not repeated.

Usage

secondorderprops(
  xiim,
  gblargs = NULL,
  covarargs = NULL,
  cencovarargs = NULL,
  paircorrargs = NULL,
  returnrotmean = FALSE
)

Value

A named list of estimated properties. When multiple estimators have been requested for the same property, then the entry in the list is itself a list, with each entry corresponding to a different estimator.

Arguments

xiim

A spatstat im object with pixel values that are either TRUE, FALSE or NA. TRUE represents foreground, FALSE represents background and NA represents unobserved locations.

gblargs

A list of named arguments passed to gblemp and gbl.cvchat. The estimators used can be specified by passing an argument named 'estimators' contain a list of estimator names, as given in gbl. If NULL then GBL will not be estimated.

covarargs

A list of named arguments passed to racscovariance.cvchat. If NULL then covariance will not be returned.

cencovarargs

A list of named arguments passed to cencovariance.cvchat. If NULL then pair correlation will not be returned.

paircorrargs

A list of named arguments passed to paircorr.cvchat. If NULL then pair correlation will not be returned.

returnrotmean

Logical. If FALSE the anisotropic estimates of covariance and pair-correlation will be returned as im objects. If TRUE then average covariance and pair-correlation over all directions will be returned as fv objects.

Warning

The user interface for this function is substantially more stretched the knowledge of the author, Kassel Hingee. Therefore, there is greater chance of encountering bugs. Kassel Hingee apologises for any bugs you encounter, and requests to be informed (thank you!).

Examples

xiim <- as.im(heather$coarse, value = TRUE,
              na.replace = FALSE)
gblargs = list(boxwidths = seq(1, 10, by = 1), estimators = c("GBLemp", "GBLcc.pickaH"))
covarargs = list(estimators = "all")
cencovarargs = list(estimators = "pickaH")
paircorrargs = list(estimators = "pickaH")
returnrotmean = TRUE
secondests <- secondorderprops(xiim,
   gblargs = gblargs,
   covarargs = covarargs,
   cencovarargs = cencovarargs,
   paircorrargs = paircorrargs, 
   returnrotmean = FALSE)