Hello Antonio,<div><br></div><div>I'll read the specs regarding JaxaPalser type, out software is reading those and they look well, but I didn't analyze the format in detail. I've also added georeferencing (Calculating the affine transform based on the 4 corners and assigning a coordinate system (WGS84 instead of ITRF97 at the moment as we are based on 1.8.0) to the JaxaPalsar driver.</div>
<div><br></div><div>For the time being, I'm writing new drivers for the PRISM and AVNIR2 sensors, based on the current JaxaPalsar driver, as we are constrained in time, but I think that once I send the code, the diffing could be useful when writing the new unified driver. I'm ready to collaborate if you think another approach could more useful. The idea is to have PRISM and AVNIR2 support ASAP.</div>
<div><br></div><div>Regards,</div><div>Rodolfo Bonnin.<br><br>2011/6/26 Antonio Valentino <<a href="mailto:antonio.valentino@tiscali.it">antonio.valentino@tiscali.it</a>>:<br>> Hi Frank,<br>><br>> Il 26/06/2011 13:42, Frank Warmerdam ha scritto:<br>
>> On 11-06-26 06:34 AM, Antonio Valentino wrote:<br>>>> Hi Frank,<br>><br>> [...]<br>><br>>>> Looking deeper into the ceos2 code it seems to me that the "recipe" for<br>>>> decoding ALOS-PALSAR products do not match at all specifications I can<br>
>>> find at<br>>>><br>>>> <a href="http://www.eorc.jaxa.jp/ALOS/en/doc/fdata/PALSAR_L10_J_ENa.zip">http://www.eorc.jaxa.jp/ALOS/en/doc/fdata/PALSAR_L10_J_ENa.zip</a><br>>>><br>>>> Does that recipe refers to another format specification document?<br>
>><br>>> Antonio,<br>>><br>>> I don't know.<br>><br>> well, in this case I guess it is not a good idea to modify that recipe<br>> in any manner<br>><br>>>> Also, regarding the ceos2 driver in general, I didn't find an obvious<br>
>>> way to support CEOS products with multiple image files (like multi<br>>>> polarization PALSAR or ALOS PRISM).<br>>><br>>> Ah, that may well have been part of Phil's rationale for handling<br>
>> multi-polarization data with custom drivers. I don't know how hard<br>>> it would be to do so within the existing SAR_CEOS driver.<br>><br>> OK, thank you for confirming my analysis<br>><br>>> Possibly you could set up a multi-polarization driver based on the<br>
>> low level ceos2 code but with a somewhat different recipe mechanism?<br>><br>> Yes I could try.<br>><br>> A new CEOS multi-band driver could be used for ALOS (all three sensors)<br>> and Landsat.<br>
> There are other sensors that could be interesting to support?<br>> Where can I find pointers and format spec?<br>><br>> ciao<br>><br>> --<br>> Antonio Valentino<br>> _______________________________________________<br>
> gdal-dev mailing list<br>> <a href="mailto:gdal-dev@lists.osgeo.org">gdal-dev@lists.osgeo.org</a><br>> <a href="http://lists.osgeo.org/mailman/listinfo/gdal-dev">http://lists.osgeo.org/mailman/listinfo/gdal-dev</a><br>
><br><br><br><br>-- <br>Ing. Rodolfo Bonnin<br>SUR Emprendimientos Tecnológicos<br><br>Perú 345 Piso 5to Oficina "B" (C1067AAG)<br>Ciudad de Buenos Aires, Argentina<br>Tel. +54 (11) 4342-2976/84<br><a href="mailto:rodolfobonnin@suremptec.com.ar">rodolfobonnin@suremptec.com.ar</a><br>
<a href="http://www.suremptec.com">www.suremptec.com</a><br><br></div>