[R-sig-Geo] Simulating raster or spatialgriddataframe images

[Edzer J. Pebesma]

Andrew,

If you want to simulate particular random fields that are Gaussian or 
binary, described by a second order properties (mean, variogram), then 
you may want to look into the simulation capabilities of packages 
RandomFields or gstat. If you want to simulate objects (circles, 
ellipses, stars), then you may need to do some point-in-polygon stuff, 
which is fairly easy to automate. Perhaps using overlay (in package sp) 
if you have the object as a SpatialPolygons[DataFrame] object.

Hth,
--
Edzer

Andrew Niccolai wrote:
> Greetings,
>
> At the risk of being ridiculed for such a general question, can anyone offer
> suggestions or decent code to simulate grid images with specific patterns?
> I have been reading R Graphics and I am not sure if a grob will suffice, but
> I would like to create a "controlled" circular or rectangular image in which
> I can insert a star pattern or a pattern of high valued clumps evenly
> distributed around the image.  I am interested in creating these controlled
> or simulated images in order to better understand the ability of gstar and
> local moran's I statistics in discriminating among image objects with
> different internal patterns but that share similar overall averages and
> standard deviations.  Can anyone suggest code or techniques inside of R that
> will allow me to create a series of SGDF objects that have a specific set of
> internal patterns associated with them.
>
> Thanks in advance and I apologize for the "vagueness" of the query.
>
> Andrew Niccolai
> Doctoral Candidate
> Yale School of Forestry
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>