Improve STATISTICS_VALID_PERCENT:

[alexisahedo]

Hello dear everyone,

Thanks a lot for all your help so far, I’ll like to continue my questions regarding the RGB topic. Using the general workflow shown in your Google Colab notebooks I made a script to process GRD images to ARD CEOS standard and finally to RGB, sadly the statistic: STATISTICS_VALID_PERCENT barely reach the 69, is there a way to optimize the valid percent value of the raster?

**
Also, I’ll like to add that I have had been working on a new method to transform the BEAM-DIMAP into a GeoTiff single band grayscale, I used as base the method of [ard_to_rgb()] in script [grd_to_ard.py], if you have interest in it I’ll gladly add it in the conversation for your check and (maybe) future add to the main ost project, it was my way around the “GDAL [to_tif]=true” problem.
Regarding this new script, I have a question about the nodata=value in the rasterio.open() function, it is set 0 by default, but without any apparent reason, the pixels with no data around the processed image are set to -130 which cause an ugly black framework with no data. Why when rasterio take the original image BEAM-DIMAP and turn it into a GeoTiff change the no data values from 0 to -130? Even thought I tried to follow the same workflow for [ard_to_rgb()] method, the no data value change. Do you have any idea why?

As always, I really appreciate all your hard work and kind help.

Saludos amigos.
Alexis A.