[gdal-dev] An issue about using GDALWARP to do the NTv2 transformation

Frank Warmerdam warmerdam at pobox.com
Wed Dec 15 21:37:55 EST 2010


On 10-12-15 09:24 PM, heng.feng wrote:
>
> http://osgeo-org.1803224.n2.nabble.com/file/n5838904/gdalwarp_problem.doc
> gdalwarp_problem.doc
>
> Hi there,
>
> I was trying to use GDALWARP command to do the NTv2 transformation on some
> German map data (raster data). However I have come across a lot of errors
> saying "failed to find a grid table for the location". I used the same grid
> shift file and the same projection settings in the ogr2ogr command to do the
> NTv2 transformation on the MapInfo map data files (vector data) we have got
> . It seems working perfectly without any error message. The grid shift file
> was downloaded from
> http://translate.google.com/translate?js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&sl=de&tl=en&u=http%3A%2F%2Fcrs.bkg.bund.de%2Fcrseu%2Fcrs%2Fdescrtrans%2FBeTA%2Fde_dhdn2etrs_beta.php
> this link .
>
> I am not sure whether I have used the GDALWARP command correctly. It seems
> that I got the error messages for all the raster map data I have got here by
> running the GDALWARP command to do the NTv2 transformation on them. I have
> attached two screenshots for your reference. As follows, please find the
> GDAL command I was using and some the error messages copied from the FWTools
> shell window. Note that I have turned on the debug mode before running the
> command in FWTools shell window:
...
> pj_open_lib(C:\fh\comarch\BETA2007.gsb): call
> fopen(C:\fh\comarch\BETA2007.gsb)
> - succeeded
> NTv2 DHDN90   62x84: LL=(5.5,47) UR=(15.6666667,55.3)
> pj_apply_gridshift(): failed to find a grid shift table for
>                        location (5.0347670dW,55.2216906dN)
>     tried: C:\fh\comarch\BETA2007.gsb


Heng,

The BETA2007.gsb file is valid for the area LL=(5.5,47) UR=(15.6666667,55.3)
but you are trying to warp points like 5degrees *west* 55 north.  Note that
5d W is -5 degrees and is roughly 10 degrees away from the closest area
in the grid shift file.

Does it seem possible that somehow your raster data has been incorrectly
georeferenced with the longitude sign wrong?  If not, then it seems likely
your raster file extends outside the region in the grid shift file.  Does
this seem reasonable?  Or are you convinced that your file ought to all
be within the area of the grid shift file?

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent



More information about the gdal-dev mailing list