[GRASS-user] integration of sample.c algorithms into r.resample

[Dylan Beaudette]

On 8/14/06, Glynn Clements <glynn at gclements.plus.com> wrote:
>
> Dylan Beaudette wrote:
>
> > How about this:
> > 1. keep r.resample and update the documentation accordinly.
> > 2. toss r.bilinear, possibly leave a note somewhere in the docs
>
> Or replace it with a script which calls "r.interpolate method=bilinear".
>
> > 3. create a new module 'r.resample.3' (?) for the three methods that it would
> > do? or 'r.interpolate' (?)
>
> I prefer r.interpolate.

This works for me.

>
> > Note: the r.interpolate name would make sense for moving from larger to
> > smaller grid spacing ( 100m -> 10m), however it wouldn't make as much sense
> > for moving from smaller to larger grid spacing (10m -> 100m).
>
> Even in that case, it will normally still perform interpolation. The
> only case where no interpolation occurs is when the new resolution is
> an exact multiple of the old resolution and the alignment is correct.
>

I am working up a test case for this, cobbling together an
'r.interpolate' via a series of commands to show the differences to
r.resample.

> > In this case
> > something like r.aggregate  would make more sense. However this potentially
> > opens a new can of worms- as there are many ways to aggregate cells... which
> > might call for a new vector / raster module.
>
> An aggregation tool would be useful in itself. At present, the
> only solutions which I know of involve either r.neighbors or
> r.average, both of which are a bit awkward.
>
> --
> Glynn Clements <glynn at gclements.plus.com>
>

I will post my results a little later today,

Cheers,

Dylan