raster (version 1.6-16)

Sample: Random or regular sample

Description

Take a random or systematic sample the values from a Raster object.

Usage

sampleRandom(x, size, ...)
sampleRegular(x, size, extent=NULL, cells=FALSE, asRaster=FALSE, corners=FALSE)

Arguments

x

a Raster object

size

positive integer giving the number of items to choose.

...

Additional arguments: na.rm, extent, and cells. If TRUE NA values are removed from random sample

extent

Extent. To limit regular sampling to the area within that box

cells

Logical. Also return sampled cell numbers (if asRaster==FALSE)

asRaster

Logical. If TRUE, a rasterLayer is returned, rather then the sampled values

corners

Logical. If TRUE, the extreme corner cells are always included. This makes for rather irregular sampling when n is small; but it can be useful when asRaster=TRUE, to maintain the same extent as the input RasterLayer

Value

• A vector, matrix (if cells=TRUE; or for a multi-layered object), or RasterLayer (if asRaster=TRUE)

Details

With na.rm=TRUE, the returned sample may be smaller than requested

Examples

r <- raster(system.file("external/test.grd", package="raster"))
 v <- sampleRegular(r, size=100)
 s <- stack(r, r)
 sampleRandom(s, size=10, cells=TRUE)