RGB profile for pconvert with multiple input files

Hi,

I want to use pconvert in order to generate quicklooks from processed S1 GRD products. I have two TIFF files, one for each of VV and VH polarizations. I tried to use pconvert as follows:

pconvert -p /path/to/vv_vh_profile.rgb -f png -o /path/to/quicklooks/ /path/to/vv.tif/ /path/to/vh/tif

but I don’t know how to write the RGB profile. From the GUI I got the following file:

name=vv_vh_profile
blue=Intensity_VV/Intensity_VH
green=Intensity_VH
red=Intensity_VV

How can I select one band from each input file?

(I guess I could use a single file with the two polarizations in it but I’d like to keep them in separate files.)

Thank you!

Unfortunately pconvert cannot handle two input files. You need to merge both polarisations into one file.

Thanks for your answer.

However I don’t understand how RGB profiles should be written. Once my two tiffs are merged into one tiff containing band 1 and band 2, how can I select them in the profile file? I tried setting VV and VH as band description with GDAL and using them in the profile but I got an “Undefined symbol” error.
I don’t understand what Intensity_VV stands for in the profile generated by the GUI since this symbol is not in the manifest.safe nor it is returned by gdalinfo.

Have you tried to open the merged product in SNAP and create an RGB image?
While defining the RGB image you can store the profile file.

I tried it and it works with SNAP Desktop. However when I do it in command line, it fails with this error:

reading file tmp_merged.tif
INFO: org.hsqldb.persist.Logger: dataFileCache open start
loading RGB profile from ‘/data/vv_vh_profile.rgb’…
creating histogram for band ‘band_1’…
creating histogram for band ‘band_2’…
creating histogram for band ‘virtual_blue’…
creating RGB image…
error: failed to create image: null

I found this related post but it doesn’t solve the issue (I set Xmx44G and the result is still the same).

Hey,
Im having the same error, reported in this post:

Maybe it helps finding a solution to the issue.
Thanks for the ongoing support.