Yilmaz Arslanoglu wrote:
I was considering too
to utilize
the Natural Neighbours concept, not only for nearest neighbour
identification, but also for
interpolating the values attributed to spatial points.
If you google for "triangulation" and "Delaunay" you'll find lots of
references and code.
One good one to take a look at is Jonathan Shewchuk's "triangle"
http://www.cs.cmu.edu/~quake/triangle.html
It is fast, robust, and pretty easy to use -- it is a command line
program, and can also be compiled as a library. I think it will do
linear interpolation of values to new points for you.
It may have licensing issues -- it one of those annoying "free for
academic use" clauses, but no standard open source license.
You also may want to take a look at ucar's ntagrid:
http://ngwww.ucar.edu/ngmath/natgrid/intro.html
since one needs to adapt the tesellation with the newly inserted query
point at eachnew query.
I don't think Triangle has an API for that, but it does dynamically add
points itself when you ask it to create a "quality" triangulation, so
you may be able to hook into that.
As I think about it, though, I'm not sure you need to do that. If you
are linearly interpolation among your known points (that is, fitting a
plane to the thre points of the traingle that your new point is in), you
don't need to include an added point when doing the next one -- yes it
might be a natural neighbor of a future point, but it will lie on teh
same plane, so it won't change the result. If you do a fancier
interpolation, it may make a difference.
As it happens, I need to do something very similar soon, so I'd love to
hear what you come up with.
-Chris
--
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
chris.bar...@noaa.gov
_______________________________________________
gdal-dev mailing list
gdal-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/gdal-dev
