maxnndist {spatstat.geom}

Compute Minimum or Maximum Nearest-Neighbour Distance

Description

A faster way to compute the minimum or maximum nearest-neighbour distance in a point pattern.

Usage

minnndist(X, positive=FALSE, by=NULL)
maxnndist(X, positive=FALSE, by=NULL)

Arguments

X

A point pattern (object of class "ppp").

positive

Logical. If FALSE (the default), compute the usual nearest-neighbour distance. If TRUE, ignore coincident points, so that the nearest neighbour distance for each point is greater than zero.

by

Optional. A factor, which separates X into groups. The algorithm will compute the distance to the nearest point in each group.

Details

These functions find the minimum and maximum values of nearest-neighbour distances in the point pattern X. minnndist(X) and maxnndist(X) are equivalent to, but faster than, min(nndist(X)) and max(nndist(X)) respectively.

The value is NA if npoints(X) < 2.

Value

A single numeric value (possibly NA).

If by is given, the result is a numeric matrix giving the minimum or maximum nearest neighbour distance between each subset of X.

Author(s)

Adrian Baddeley [e-mail], Rolf Turner [e-mail] and Ege Rubak [e-mail].

See Also

nndist

Examples

  min(nndist(swedishpines))
  minnndist(swedishpines)

  max(nndist(swedishpines))
  maxnndist(swedishpines)

  minnndist(lansing, positive=TRUE)

  if(interactive()) {
     X <- runifrect(1e6)
     system.time(min(nndist(X)))
     system.time(minnndist(X))
  }

  minnndist(amacrine, by=marks(amacrine))
  maxnndist(amacrine, by=marks(amacrine))