[gdal-dev] gdalbuildvrt problem with heterogenous band characteristics

Even Rouault even.rouault at mines-paris.org
Thu Aug 16 07:32:40 PDT 2012


Selon Margherita Di Leo <diregola at gmail.com>:

> Hi All,
>
> On Mon, Jul 30, 2012 at 7:31 PM, Casper Børgesen (CABO) <CABO at niras.dk>wrote:
>
> >  Hi! ****
> >
> > ** **
> >
> > I’m trying to create a VRT file from 4 GeoTIFFs using gdalbuildvrt. Three
> > of the GeoTIFFs are created using OrthoVista and the last one created with
> > an earlier version of the same program. When I run gdalbuildvrt I get the
> > following error message when I try to add the file created by the early
> > version: ****
> >
> > ** **
> >
> > Warning 6: gdalbuildvrt does not support heterogenous band
> > characteristics. Skipping...****
> >
> > ** **
> >
> > I get the same error message, so may be related... I've tried to make a
> virtual mosaic, using SRTM DEMs,
>
> gdalbuildvrt -input_file_list list.txt STM_CP-DEMS_EU_mosaic.vrt
>
> and got:
>
> 0...Warning 6: gdalbuildvrt does not support heterogenous band
> characteristics. Skipping STM_CP-DEMS_2500015000.tif
> 10.Warning 6: gdalbuildvrt does not support heterogenous band
> characteristics. Skipping STM_CP-DEMS_2500025000.tif
> ..Warning 6: gdalbuildvrt does not support heterogenous band
> characteristics. Skipping STM_CP-DEMS_2500035000.tif
> [....]
>
> Hence, I made my mosaic using:
>  gdalwarp --optfile list.txt STM_CP-DEMS_EU_mosaic.tif
> It took a bit more time obviously, but it didn't complain.
>
> So.. Could it be a bug of gdalbuildvrt, or.. ?

I don't think it is a bug. gdalbuildvrt is a "lightweight" but restrictive
mosaicing solution, whereas gdalwarp is "heavier" but more capable.

gdalbuildvrt makes various checks to ensure that the tiles to be mosaiced are
very similar. Perhaps some of the tests are too restrictive. You should try to
compare the output of gdalinfo between a file that is integrated into the VRT
and one that is skipped to see which differences there are (band data type,
presence/absence of color table, color interpretation, etc...)

>
> Thanks and regards,
>
>
> --
> Dr. Margherita Di Leo
>




More information about the gdal-dev mailing list