[Gdal-dev] Problems with RS2 driver for import to grass

Benjamin Deschamps bdesc100 at gmail.com
Wed Mar 18 16:23:13 EDT 2009


Phil,

I am using GDAL 1.6.0-5 and GRASS 6.4.0RC3. I guess my question was  
more whether it was possible to keep the data in complex format, with  
two layers per channel for imaginary and real components. I'm having a  
hard time getting the image to look and have the values of a "real"  
radarsat-2 image... I don't understand how the data can be used in  
integer format (-1,0,1).

Benjamin


On 18-Mar-09, at 4:06 PM, Philippe Vachon wrote:

> Hi Ben,
>
> I can't recall the exact details of how the GRASS r.in.gdal mechanism
> works, but the results you describe are similar to importing  
> calibrated
> data into an integer raster. Calibrated SAR data is typically
> represented as floats, as many of the values are in [0, 1), if I am  
> not
> mistaken. As such, the values when cast to integers will be clamped  
> to 0
> or 1 (or -1, in the case of the complex values).
>
> I'm not certain about the GCP issue, since I can import all of the  
> GCPs
> in the Altona dataset using a little app I just whipped up. Perhaps it
> could be something specific to GRASS that I'm not aware of. Happen to
> know what version of GDAL you're using?
>
> Cheers,
> Phil
>
> On Wed, Mar 18, 2009 at 12:48:01PM -0700, Benjamin Deschamps wrote:
>>
>> I'm having a few issues importing RADARSAT-2 data to GRASS. I used:
>>
>> r.in.gdal
>> input=RADARSAT_2_CALIB:SIGMA0:/Users/Benjamin/Desktop/ 
>> PK6621_DK406_FQ9_20080405_124900_HH_VV_HV_VH_SLC_(Altona_Promo)/ 
>> product.xml
>> output=altona
>>
>> to import the sigma nought calibrated data. For some reason this only
>> imports 31 GCPs out of 350, and the values of the resulting GRASS  
>> layer are
>> limited to -1, 0 and 1, so the map doesn't look like anything
>> understandable. If I import uncalibrated values instead
>> (RADARSAT_2_CALIB:UNCALIB:etc) I get a much more appropriate range of
>> values, and all 350 GCPs are included.
>>
>> How can I get all GCPs and a more appropriate range of values for  
>> the sigma
>> nough subdataset? The dataset is available as part of the  
>> RADARSAT-2 demo
>> dataset at
>> ftp://ftp.mda.ca/Fine%20Quad-Pol%20Dataset/Altona_%20R2_FineQuad9_HH_VV_HV_VH_SLC.zip
>> ftp://ftp.mda.ca/Fine%20Quad-Pol%20Dataset/Altona_%20R2_FineQuad9_HH_VV_HV_VH_SLC.zip
>>
>> Thanks, Benjamin
>> -- 
>> View this message in context: http://n2.nabble.com/Problems-with-RS2-driver-for-import-to-grass-tp2499302p2499302.html
>> Sent from the GDAL - Dev mailing list archive at Nabble.com.
>>
>> _______________________________________________
>> gdal-dev mailing list
>> gdal-dev at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/gdal-dev



More information about the gdal-dev mailing list