[R-sig-Geo] nearest distance in matrix

[Barry Rowlingson]

Marius Gilbert wrote:
> Hello,
> 
> I'm trying to use R to simulate biological invasions, and got stuck with 
> the following:
> 
> Is there a function that uses a matrix of occupied/empty values (1 or 0) 
> as input, and producing an output matrix of the same size, with each cell 
> containing the distance to the nearest occupied cell (1) of the input 
> matrix ?

  Try this, which uses an internal, undocumented Spatstat function 
'exactPdt':

library(spatstat)
nmatdist <- function(aMat){
   storage.mode(aMat) <- "logical"
   win <- owin(xrange=c(1,ncol(aMat)),yrange=c(1,nrow(aMat)),mask=aMat)
   m <- exactPdt(win)
   return(m$d)
}

> For example, the input matrix:
> 
> 0    1    1    0
> 0    0    1    0
> 0    0    0    0
> 0    0    0    0
>

> Would produce an output matrix like:
> 
> 1    0    0    1
> 1.4  1    0    1
> 2.2  1.4  1    1.4
> 2.8  2.2  2.0  2.2	

  test:

 > m1
      [,1] [,2] [,3] [,4] [,5]
[1,]    0    0    0    0    0
[2,]    1    0    0    0    0
[3,]    1    1    0    0    0
[4,]    0    0    0    0    0
 > nmatdist(m1)
      [,1]     [,2]     [,3]     [,4]     [,5]
[1,]    1 1.414214 2.236068 2.828427 3.605551
[2,]    0 1.000000 1.414214 2.236068 3.162278
[3,]    0 0.000000 1.000000 2.000000 3.000000
[4,]    1 1.000000 1.414214 2.236068 3.162278

  looks about right!

Baz