I am stuck with a AAI grid export problem (r.out.gdal input=NAME
format=AAIGrid type=Byte output=NAME) and I have no clue if it is a gdal
problem or a GRASS (r.resamp.*) problem.
1) Import dataset (r.in.gdal) - export it straight afterwards (r.out.gdal see
above) - both raster images fit onto each other using QGIS
2) Import dataset (r.in.gdal) - applying r.resamp.stats (90m res in 1000m
res), export with r.out.gdal (see above) - new image has a shift of +/- 100
Km to the south-east compared to the original image
I assume its a r.resamp. problem but when I check the header files
(row,col,res.,proj, etc.) of both exported files, they are the same which
suprises me because they don't fit on each other.
have you changed g.region to your new settings before export?
Wolfgang
Martin Wegmann schrieb:
Hello,
I am stuck with a AAI grid export problem (r.out.gdal input=NAME format=AAIGrid type=Byte output=NAME) and I have no clue if it is a gdal problem or a GRASS (r.resamp.*) problem.
1) Import dataset (r.in.gdal) - export it straight afterwards (r.out.gdal see above) - both raster images fit onto each other using QGIS
2) Import dataset (r.in.gdal) - applying r.resamp.stats (90m res in 1000m res), export with r.out.gdal (see above) - new image has a shift of +/- 100 Km to the south-east compared to the original image
I assume its a r.resamp. problem but when I check the header files (row,col,res.,proj, etc.) of both exported files, they are the same which suprises me because they don't fit on each other.
On Thursday 30 November 2006 13:43, Wolfgang wrote:
have you changed g.region to your new settings before export?
I changed g.region res. during the r.resamp.stats but for export of the two
raster g.region wasn't changed
BTW r.info of the two raster gives same res/coord. for both.
Martin
Wolfgang
Martin Wegmann schrieb:
> Hello,
>
> I am stuck with a AAI grid export problem (r.out.gdal input=NAME
> format=AAIGrid type=Byte output=NAME) and I have no clue if it is a gdal
> problem or a GRASS (r.resamp.*) problem.
>
> 1) Import dataset (r.in.gdal) - export it straight afterwards (r.out.gdal
> see above) - both raster images fit onto each other using QGIS
>
> 2) Import dataset (r.in.gdal) - applying r.resamp.stats (90m res in 1000m
> res), export with r.out.gdal (see above) - new image has a shift of +/-
> 100 Km to the south-east compared to the original image
>
> I assume its a r.resamp. problem but when I check the header files
> (row,col,res.,proj, etc.) of both exported files, they are the same which
> suprises me because they don't fit on each other.
>
> Any idea where to investigate further?
>
> regards, Martin
>
> _______________________________________________
> grassuser mailing list
> grassuser@grass.itc.it
> http://grass.itc.it/mailman/listinfo/grassuser
I tested various r.out.gdal settings (-r is a very handy function! - thanks
Hamish) and found no errors using various files.
Finally I ended up doing further test with r.resamp.stats where I found the
problem.
Hence r.out.gdal shows no inconsistency but r.resamp.stats - I'll post another
message concerning this behaviour in a few minutes.
Martin
On Friday 01 December 2006 01:51, Hamish wrote:
Martin Wegmann wrote:
> On Thursday 30 November 2006 13:43, Wolfgang wrote:
> > have you changed g.region to your new settings before export?
>
> I changed g.region res. during the r.resamp.stats but for export of
> the two raster g.region wasn't changed
>
> BTW r.info of the two raster gives same res/coord. for both.
what about with using the new region sensitive GDAL output? (resamples)
"r.out.gdal -r"
On Friday 01 December 2006 08:34, Martin Wegmann wrote:
I tested various r.out.gdal settings (-r is a very handy function! - thanks
Hamish) and found no errors using various files.
Finally I ended up doing further test with r.resamp.stats where I found the
problem.
Hence r.out.gdal shows no inconsistency but r.resamp.stats - I'll post
another message concerning this behaviour in a few minutes.
Martin
On Friday 01 December 2006 01:51, Hamish wrote:
> Martin Wegmann wrote:
> > On Thursday 30 November 2006 13:43, Wolfgang wrote:
> > > have you changed g.region to your new settings before export?
> >
> > I changed g.region res. during the r.resamp.stats but for export of
> > the two raster g.region wasn't changed
> >
> > BTW r.info of the two raster gives same res/coord. for both.
>
> what about with using the new region sensitive GDAL output? (resamples)
> "r.out.gdal -r"
>
>
> r.in.gdal out=NAME
> g.region rast=NAME
> r.out.gdal -r in=NAME
>
>
> ?
> Hamish
hmm, interesting - I assume it is the same behaviour and due to the larger
area I am working on, my shift is more pronounced.
Attached a jpg of r and elevation.dem correlation (Moving Window size 3x3
multiplied with 1000) - the deviation from 1000 (or corr. val 1) is just
present in parts of the image - I would have assumed that the whole image has
a constant deviation from 1000, hence there might be another error.
Martin
Cheers,
Dylan
On Friday 01 December 2006 08:34, Martin Wegmann wrote:
> I tested various r.out.gdal settings (-r is a very handy function! -
> thanks Hamish) and found no errors using various files.
> Finally I ended up doing further test with r.resamp.stats where I found
> the problem.
> Hence r.out.gdal shows no inconsistency but r.resamp.stats - I'll post
> another message concerning this behaviour in a few minutes.
>
> Martin
>
> On Friday 01 December 2006 01:51, Hamish wrote:
> > Martin Wegmann wrote:
> > > On Thursday 30 November 2006 13:43, Wolfgang wrote:
> > > > have you changed g.region to your new settings before export?
> > >
> > > I changed g.region res. during the r.resamp.stats but for export of
> > > the two raster g.region wasn't changed
> > >
> > > BTW r.info of the two raster gives same res/coord. for both.
> >
> > what about with using the new region sensitive GDAL output? (resamples)
> > "r.out.gdal -r"
> >
> >
> > r.in.gdal out=NAME
> > g.region rast=NAME
> > r.out.gdal -r in=NAME
> >
> >
> > ?
> > Hamish
>
> _______________________________________________
> grassuser mailing list
> grassuser@grass.itc.it
> http://grass.itc.it/mailman/listinfo/grassuser
