this is currently discussed in several topics
but we haven’t found the reason yet.
O, then sorry for dublicating
Didn’t perform search in these topics because there is actually no error=)
First i even thought there is a problem with my pc, but i tried anoter one with another OS (MAC), repeated the whole workflow but with lesser number of images (that MAC had much weaker hardware) - the same problem but another percent value
Then downloaded another small group of images of the same objects - and everything went fine! Somehow initial data matters, im not good enough to understand in what exact way
Thanks for the answer!
no worries, just wanted to put you on the latest status.
We wondered if the changed location of SRTM 3Sec has caused this, but the more we think about it, the less it makes sense. On the other side, some have reported that switching to SRTM 1Sec during the interferogram formation has solved their case. But as the actual export does not re-download elevation data, it doesn’t make entirely sense to me.
1 Like
Hi. I have faced several problems, which I have been trying to solve, the first has been the export:
1st usea snap2stamp, this gave me problems, so on the recommendation of @mdelgado I did it in the GUI.
Using SNAP directly, the export seems to work.
So I keep running:
source /home/igncio/StaMPS-4.1-beta/StaMPS_CONFIG.bash
here I get another error,
according to the recommendations consulted in this forum, I ran snap2stamps again from the beginning only with the conflicting images … this did not solve the problems of the interferometry
since the interferograms were badly generated again
So I opt for the second recommendation which is to remove them from the folder PATCH_1.
and continue now if with Stamps in Matlab environment, I execute the first step and I get the following error:
I check the folders and pscands is actually empty, so I don’t know how to continue.
Any recommendation or clarification of why interferograms are generally wrong and why they do not generate pscandidats?
all indicates that the Coregistration was not successful.
Can ypu please check an image pair and visualize if it contains information of both dates?
The rest of the interferograms, it created them properly.
2 random examples:
Except in this pair, which I don’t understand why directly it doesn’t form it.
Is it possible that the problem in Stamps is due to the export?
in any case, when exporting directly in the snap GUI
according to this topic (EAP phase correction not possible) the error on EAP Phase correction only occurs in old versions of SNAP. Are you sure you are using the latest version?
Could it be that I was confused on the update?
Should I have downloaded the program again and installed it instead of updating it directly?
yes, absolutely. The internal updates only cover updates within one version, but will not update from version 6 to version 8.
Hi, I finally reinstalled 8. But once again an error has occurred, in this case:
I know there is a solution from Error in SAR image corregistration: Error: [Nodeld: CreateStack] org.jblas.NativeBlas.dgemm(CCIIID[DII[DIID[DII)V) by @thho, but in this case for version 8 of SNAP with Ubuntu 18.04.5 LTS, I am not getting results with the libgfortran3 library
hi, I had the same problems as in this recent post: Error in SAR image corregistration: Error: [Nodeld: CreateStack] org.jblas.NativeBlas.dgemm(CCIIID[DII[DIID[DII)V)
even if you install libgfortran3 it doesn’t work for SNAP 8
The solution that does work is to go back to previous versions.
However, this conflicts with the idea of updating to the latest version to avoid conflict with SRTM.
So. Any guidelines on which version to use for SNAP-Stamps?
At this point I feel frustrated that I cannot go any way.
please also install libgfortran5, this has reportedly solved the issue here: Please test the SBAS_snap2stamps
Thank you very much for the answer, this easily solved the problem.
One question, I am using Ubuntu 18.04, but I think it is time to update it to Ubuntu 20.04.1 LTS, however I am afraid that the whole set of programs (SNAP, snap2stamps, matlab, train, etc) is not yet ready for it. Therefore, do you think it is feasible or advisable to do this update already?
sorry, I haven’t tested different settings - maybe someone who has updated Ubuntu can report here?
STAMPS: Will process patch subdirectories
STAMPS: ########################################
STAMPS: ################ Step 1 ################
STAMPS: ########################################
STAMPS: Directory is PATCH_1
PS_LOAD_INITIAL_GAMMA: Starting
PS_LOAD_INITIAL_GAMMA: Loading data into matlab…
master_master_flag =
'0'
READPARM: heading:=190.26904863170714
SETPARM: heading = 190.269
READPARM: radar_frequency:=5.40500045433435E9
SETPARM: lambda = 0.055466
READPARM: sensor:=SENTINEL-1A
SETPARM: platform = SENTINEL-1A
READPARM: range_pixel_spacing:=2
READPARM: near_range_slc:=901175.0317591032
READPARM: sar_to_earth_center:=7073347.970017257
READPARM: earth_radius_below_sensor:=6367489.32120738
READPARM: center_range_slc:=901175.0317591032
READPARM: azimuth_lines:=6882
READPARM: prf:=1717.128973878037
Index in position 2 exceeds array bounds.
Error in ps_load_initial_gamma (line 108)
rg=rgn+ij(:,3)*rps;
Error in stamps (line 263)
ps_load_initial_gamma;
Still i’m getting this error after re-installation of Matlab
And i replaced the ps_load_initial_gamma.m file also.
Please tell me is there any solution for this error?
hi,
I did the snap2stamps preprocessing and stamps processing using Ubuntu 20.04, and it worked well. i used SNAP version 8, executed snap2stamps with python 2.7 and used stampsV4.1 . i don’t recall having major problems with the processing.
Good luck
1 Like
Hello,
I think you have this error cause your working directory on matlab should be the “INSAR_…” folder not the “PATCH_1” folder. try to re-lunch matlab from your INSAR folder and re-run stamps.
I hope this was helpful.
Good luck
ABraun solved my error. Thank you for your response.
Software will calculate the values according to yearly basis. If you have imagery for 5 years then software will divide the result by 5 to get yearly value and if you work on 6 month then it will divide the value by 0.5 so it gives the results in yearly basis.
I hope it helped.
this is an approximation for shorter periods but based on the assumption that the deformation rate is constant over the investigated period.
1 Like