[gdal-dev] GDAL translate 16 bit to 8 bit

[Hanlie Pretorius]

Hi Even,

Yes, I quoted the exact message, but it displays in the QGIS interface.

I tried gdal_translate with the -ot and -scale parameters on the
multispectral image and that worked much better, thanks.

Regards
Hanlie

2014-08-21 20:33 MGT+02:00, Even Rouault <even.rouault at spatialys.com>:
> Le jeudi 21 août 2014 08:57:32, Hanlie Pretorius a écrit :
>> Hi,
>>
>> I'm working with a 16 bit SPOT 6 image that I want to convert from 16
>> bit pixel depth to 8 bit pixel depth. I say 16 bit, but if I load the
>> file into Envi, it reports that uses only 12 bits. However gdalinfo
>> reports 16 bit (see below).
>>
>> I've manage to do so on the multispectral image, but the result is
>> poor colour wise.
>>
>> So I split the image into its four bands that is, I now have one image
>> per
>> band.
>>
>> When I now try to convert to 8 bit, I get a message that says:
>> "Warning 742327101_R1C1_band1_8bit.TIF not created".
>
> Are you quoting the exact message ? I can't find any trace of it in the
> relevant parts of the source code. Did you check that you have write
> permissions in the directory where you launch the command ?
>
> Anyway for what you want to do you likely need to explore the -scale option
> of
> gdal_translate. But that's unrelated to the error
>
>>