patchscan: Identify patches in classified raster data (either a matrix...

In ethanplunkett/gridprocess: Package for processing raster data

Identify patches in classified raster data (either a matrix or a grid object)

Description

This function takes a matrix or grid object in which cover type is classified and returns an object of the same class (grid or matrix) containing the patch id of each cell.

Usage

patchscan(x, nr = 8, na.value = -9999)
## S3 method for class 'grid'
patchscan(x, nr = 8, na.value = -9999)
## S3 method for class 'matrix'
patchscan(x, nr = 8, na.value = -9999)

Arguments

x	Either a matrix or a grid object.
nr	The neighbor rule either 4 or 8 indicating that either the 4 orthogonal neighboring cells or the 8 orthogonal and diagonal neighboring cells are considered neighbors.
na.value	The value used to encode NA's when passing data to C.

Details

#

Value

patches	An object of the same class as x containing the patch id of every cell in the landcape.
count	A count of the number of patches in the landcape.

Author(s)

Ethan Plunkett

See Also

grid-class

Examples

m <- matrix(sample(1:4, 400, replace=TRUE) , 20, 15)
g <- list(m=m, nrow=nrow(m), ncol=ncol(m), xll=0, yll=20, cellsize=1)
class(g) <- c("grid", class(g))

# With default arguments
res <- patchscan(g)

# With NA's in grid
g$m[2:3, 4] <- NA
res <- patchscan(g)
plot(res$patches, na.col="red")

# Using 4 neighbor rule (8 is the default)
res <- patchscan(g, nr=4)
plot(res$patches)

ethanplunkett/gridprocess documentation built on Feb. 1, 2024, 2:24 a.m.