[prev in list] [next in list] [prev in thread] [next in thread] 

List:       postgis-users
Subject:    Re: [postgis-users] [Raster] MNT 250 IGNF data import problem
From:       Yves Jacolin <yves.jacolin () camptocamp ! com>
Date:       2013-10-23 8:23:56
Message-ID: 1992089.AXjZ12Fl3W () tatras
[Download RAW message or body]

Thank to all for you answer.

I find the issue and now I have a nice raster mnt in my QGIS :p

For those working with a 1.10 gdal release you should use "LC_ALL=C qgis" to 
force QGIS (and GDAL) to use dot as decimal separator. By default in 1.10 the 
one of the system is used so all french user can't see raster layer in QGIS.

If you are using trunk gdal, it should work as expected.

Thanks to EvenR for giving me the trick about the gdal issue.

Y.
Le mardi 22 octobre 2013 22:19:33 Tumasgiu Rossini a écrit :
> Hi,
> 
> Some tiles generated from the MNT file might be filled with nodata value
> (-9999) so I guess that explains some "anormal" statistics values...
> You should execute the statistics statement against all tile to check if
> all statistics are the same...
> I suggest you to test some commands from your home-compiled GDAL (gdalinfo
> -hist for example) too.
> 
> Also, data from MNT files seems to be integer values, I don't think that
> the correct pixel depth is 32BF.
> 
> 
> For the QGIS issue, I think that mnt data is improper for display and
> should be processed into "displayable" raster data, such as GrayScale or
> TrueColors Bitmap. To do so, you might consider use st_colormap().
> 
> 
> 
> 
> 
> 2013/10/22 Yves Jacolin <yves.jacolin@camptocamp.com>
> 
> > Hello,
> > 
> > Yes it uses 32BUI (as far I can remember). I will check tomorrow and
> > made more test.
> > 
> > Thanks,
> > 
> > Y.
> > 
> > 2013/10/22 Pierre Racine <Pierre.Racine@sbf.ulaval.ca>:
> > >> Because I did a quick test and copied past previous SQL requests and I
> > 
> > did
> > 
> > >> not
> > >> clean it before! Of course you can remove the next 2 bands which don't
> > >> exist
> > >> but the first one should have correct statistic.
> > >> 
> > >> Thanks Pierre to let me know the logical ;)
> > > 
> > > Sorry I thought you were complaining about the error message not about
> > 
> > the stats....
> > 
> > > MNT250_L93_FRANCE.ASC is a text file right? I would suspect a bad
> > 
> > interpretation of the decimal separator.
> > 
> > > Check also what pixeltype was used by the importer with
> > 
> > ST_BandPixelType() or ST_BandMetadata(). I suspect it used 32BUI (because
> > it does not recognize the separator) when it should use 32BF.
> > 
> > > Pierre
> > > _______________________________________________
> > > postgis-users mailing list
> > > postgis-users@lists.osgeo.org
> > > http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
> > 
> > --
> > Responsable Formation et Support
> > Camptocamp France SAS
> > Savoie Technolac, BP 352
> > 73377 Le Bourget du Lac, Cedex
> > 
> > Tel (France) : +33 4 79 26 57 98
> > Tel (Suisse) : 021 619 10 43
> > Mob. : +33 6 18 75 42 21
> > Fax : 04 79 70 15 81
> > 
> > Mail : yves.jacolin@camptocamp.com
> > http://www.camptocamp.com
> > _______________________________________________
> > postgis-users mailing list
> > postgis-users@lists.osgeo.org
> > http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
-- 
Responsable Formation et Support
Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac, Cedex
Tel (France) : +33 4 79 26 57 98
Tel (Suisse) : 021 619 10 43 (new)
Mob. : +33 6 18 75 42 21
Fax : 04 79 70 15 81
Mail : yves.jacolin@camptocamp.com
http://www.camptocamp.com

_______________________________________________
postgis-users mailing list
postgis-users@lists.osgeo.org
http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
[prev in list] [next in list] [prev in thread] [next in thread]