Finds the topologically-connected components of a point pattern, when all pairs of points closer than a threshold distance are joined.

```
# S3 method for ppp
connected(X, R, …)
```# S3 method for pp3
connected(X, R, …)

X

A point pattern (object of class `"ppp"`

or `"pp3"`

).

R

Threshold distance. Pairs of points closer than `R`

units apart
will be joined together.

…

Other arguments, not recognised by these methods.

A point pattern, equivalent to `X`

except that the points
have factor-valued marks, with levels corresponding to the
connected components.

This function can be used to identify clumps of points in a point pattern.

The function `connected`

is generic. This file documents the methods for
point patterns in dimension two or three (objects of class `"ppp"`

or
`"pp3"`

).

The point pattern `X`

is first converted into an abstract graph
by joining every pair of points that lie closer than `R`

units
apart. Then the connected components of this graph are identified.

Two points in `X`

belong to the same connected component if they
can be reached by a series of steps between points of `X`

,
each step being shorter than `R`

units in length.

The result is a vector of labels for the points of `X`

where all the points in a connected component have the same label.

# NOT RUN { Y <- connected(redwoodfull, 0.1) if(interactive()) { plot(Y, cols=1:length(levels(marks(Y))), main="connected(redwoodfull, 0.1)") } X <- osteo$pts[[1]] Z <- connected(X, 32) if(interactive()) { plot(Z, col=marks(Z), main="") } # }