[Proj] Custom projection wildly inaccurate

[Martin Vermeer]

On Mon, 2006-11-20 at 08:36 -0500, Greg Troxel wrote:
>   > There seem to be several approximations. Most straight (read: 
>   > what you can do per proj/cs2cs command line alone) projective 
>   > approximations are bad.
>   > There is really only one way to do it correctly: use the 
>   > official polynomial solution.
> 
>   Yes, but that's not easily translatable for use in Proj-based
>   applications like GDAL and MapServer (as far I know, at least).
>   We will most likely have to convert the data manually, but in
>   the meantime it would be nice to have a bad-and-fast projection
>   for demonstration and testing purposes.
> 
> One thing that would work, but perhaps set a bad precedent, is to
> define a new projection in the proj source, that knows the
> polynomials.  This would be a particular transform, rather than a
> generic projection with parameters.  Alternatively, there could be a
> polynomial transform with vast numbers of parameters.

Alternatively, generate a shift grid containing the shifts produced by
the polynomial. This is the approach I am considering for the Finnish
kkj, which is officially represented by a multi-triangle (Delaunay)
affine transformation.

- Martin

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20061121/47145760/attachment.sig>