distfun.Rd
\name{distfun}
\Rdversion{1.1}
\alias{distfun}
\alias{distfun.ppp}
\alias{distfun.psp}
\alias{distfun.owin}
\title{
Distance Map as a Function
}
\description{
Compute the distance function of an object, and return it as a function.
}
\usage{
distfun(X, ...)
\method{distfun}{ppp}(X, ...)
\method{distfun}{psp}(X, ...)
\method{distfun}{owin}(X, ..., invert=FALSE)
}
\arguments{
\item{X}{Any suitable dataset representing a two-dimensional
object, such as a point pattern (object of class \code{"ppp"}),
a window (object of class \code{"owin"}) or a
line segment pattern (object of class \code{"psp"}).
}
\item{\dots}{
Extra arguments are ignored.
}
\item{invert}{
If \code{TRUE}, compute the distance transform of the
complement of \code{X}.
}
}
\details{
The \dQuote{distance function} of a set of points \eqn{A} is the
mathematical function \eqn{f} such that, for any
two-dimensional spatial location \eqn{(x,y)},
the function value \code{f(x,y)}
is the shortest distance from \eqn{(x,y)} to \eqn{A}.
The command \code{f <- distfun(X)} returns a \emph{function}
in the \R language, with arguments \code{x,y}, that represents the
distance function of \code{X}. Evaluating the function \code{f}
in the form \code{v <- f(x,y)}, where \code{x} and \code{y}
are any numeric vectors of equal length containing coordinates of
spatial locations, yields the values of the distance function at these
locations.
This should be contrasted with the related command \code{\link{distmap}}
which computes the distance function of \code{X}
on a grid of locations, and returns the distance
values in the form of a pixel image.
The result of \code{f <- distfun(X)} also belongs to the class
\code{"funxy"} and to the special class \code{"distfun"}.
It can be printed and plotted immediately as shown in the Examples.
A \code{distfun} object can be converted to a pixel image
using \code{\link{as.im}}.
}
\value{
A \code{function} with arguments \code{x,y}.
The function also belongs to the class \code{"distfun"} which has
a method for \code{print}.
It also belongs to the class \code{"funxy"} which has methods
for \code{plot}, \code{contour} and \code{persp}.
}
\seealso{
\code{\link{distmap}},
\code{\link{plot.funxy}}
}
\examples{
data(letterR)
f <- distfun(letterR)
f
plot(f)
f(0.2, 0.3)
}
\author{Adrian Baddeley
\email{Adrian.Baddeley@csiro.au}
\url{http://www.maths.uwa.edu.au/~adrian/}
and Rolf Turner
\email{r.turner@auckland.ac.nz}
}
\keyword{spatial}
\keyword{math}