raster (version 1.9-13)

adjacency: Adjacency

Description

Identify cells that are adjacent to a set of cells on a raster

Usage

adjacency(raster, fromCells, toCells, directions)

Arguments

raster

an object of the Raster family

fromCells

a vector of cell numbers for which adjacent cells should be calculated

toCells

a vector of cell numbers from which adjacent cells are selected

directions

in how many direction cells should be connected: 4, 8 or 16; or "Bishop"

Value

• a two column matrix with each row containing a pair of adjacent cells.

Details

Cell numbers start with 1 in the upper-left corner and increase from left to right and from top to bottom. Number of directions: 4 connects cells with one-cell rook moves, 8 with one-cell queen moves, and 16 with knight and one-cell queen moves. Bishop connects cells with one-cell diagonal moves (special cases). The function connects the outer meridians if the raster is not projeced (in a geographic (lat/lon) "projection") and there is data at longitudes -180 and 180 degrees.

Examples

r <- raster(nrows=10, ncols=10)
adj <- adjacency(raster=r, fromCells = c(1,30,55,72,100), 
				toCells = 1:ncell(r), directions=4) 
	
	
# put the result in a RasterLayer
v <- vector(length=ncell(r))
v[c(1,30,55,72,100)] <- 1
v[adj[,2]] <- 2
r <- setValues(r, v)
plot(r)


# How many time does one cell value occur next to another cell value?
r <- raster(ncol=10, nrow=10)
set.seed(0)
r[] <- round(runif(ncell(r)) * 5)
a <- adjacency(r, 1:ncell(r), 1:ncell(r), 8)
v1 <- r[a[,1]]
v2 <- r[a[,2]]
m <- matrix(0, nrow=5,ncol=5)
for(i in 1:length(v1)) {m[v1[i], v2[i]] <- m[v1[i], v2[i]] +1}
m