[R-sig-Geo] finding euclidean proximate points in two datasets

[Roger Bivand]

On Thu, 20 May 2010, Alexander Shenkin wrote:

> Hello All,
>
> I posted this over on R-help, but was then directed here.
>
> I've been pouring through the various spatial packages, but haven't come
> across the right thing yet.
>
> Given a set of points in 2-d space X, i'm trying to find the subset of
> points in Y proximate to each point in X.  Furthermore, the proximity
> threshold of each point in X differs (X$threshold).  I've constructed
> this myself already, but it's horrificly slow with a dataset of 40k+
> points in one set, and a 700 in the other.

Could you get any further with nn2() in the RANN package? I realise that 
it isn't handling distances directly, but for some k=, it might help.

Roger

>
> A very inefficient example of what I'm looking for:
>
>    X = data.frame(x=c(1,2,3), y=c(2,3,1), threshold=c(1,2,4))
>    Y = data.frame(x=c(5,2,3,4,2,5,2,3), y=c(5,2,2,4,1,2,3,1))
>    proximate=list()
>    i=1
>    for (pt in 1:length(X$x)) {
>        proximate[[i]] <- sqrt((X[pt,]$x - Y$x)2 + (X[pt,]$y - Y$y)2)
>                          > X[pt,]$threshold
>        i=i+1
>    }
>    proximate
>
> Perhaps crossdist() in spatstat is what I should use, and then code a
> comparison with X$threshold after the cross-distances are computed.
> However, I was wondering if there was another tool I should be
> considering.  David Winsemius suggested I first compare on in each
> coordinate to cull points outside the threshold on that axis first,
> before computing distances, which will help.  Any and all thoughts are
> very welcome.  Thanks in advance.
>
> Thanks,
> Allie
>

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no