NaN valyes brake stage 2

dbekaert / StaMPS

Stanford Method for Persistent Scatterers

http://homepages.see.leeds.ac.uk/~earahoo/stamps/

GNU General Public License v3.0

229 stars 121 forks source link

NaN valyes brake stage 2 #3

Closed vakhtanov closed 6 years ago

vakhtanov commented 6 years ago

I am using STAMPS on win64. Prepare SENTINEL-1 data in SNAP. There is NaN values when export from SNAP 6 to STAMPS. For the STAMPS 3.3 i download ps_load_initial_gamma from https://forum.step.esa.int/t/how-to-prepare-sentinel-1-images-stack-for-psi-sbas-in-snap-5/4981/68. There is script with corection NaN values. On pictures is error report error_stage2 error_stage2_with_print Could you help me to repair this error?

dbekaert commented 6 years ago

To eliminate further. 1) Does your xy grid in ps1 have any nan values in it? 2) When you load ps1 and hgt1 and then plot xy(:,3),xy(:,1),hgt does it look reasonable (spatial pattern and range of values)?

vakhtanov commented 6 years ago

Yes it have NaN.
Values look normal. hgt values has range -45, -38 and -32768 as Nan values. see attachment

dbekaert commented 6 years ago

From this I believe that you might have some mall-formed input data. Do your lonlat data inputted from SNAP have NaN's/ data you are not expecting in there?

vakhtanov commented 6 years ago

After export from SNAP files with data LAT, LON and Elevation get holes with NaN values like on picture lat_lon_elev

dbekaert commented 6 years ago

I included a fix for our code to remove the nan values. Please re-runs step 1 and step 2.

Though recommend you work with the SNAP team to fix this in their processor as there should not be such holes in the the lon-lat data. With the fix included above they will be removed in the stamps processing.

vakhtanov commented 6 years ago

Ok. Thank you for help!

vakhtanov commented 6 years ago

Hello! I have got error agane error3_1 here is printscrin with size of elements error3_2

dbekaert commented 6 years ago

This has been solved, in current github version.