Do you have any reference approves this issue please?
For example “Study of the impact of Polarizations for DInSAR Displacement Measures”, Parizzi et al. 2014.
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6856747
1 Like
@FeiLiu Do you change any parameter before redoing the export step? I am dealing with a stack that is resisting to show the right values…
I had the same issue and indeed, reexporting the stack the baselines were alright. Strange but true.
Hi Annamaria,
Sorry for disturbing you again.
I’m trying to get the Lon and Lat file from step 9. But it will take very long time to get the lon and lat band by using band math to add it, until now it may run over 10 hours and haven’t finished.
So, I think I could get the lon and lat bands by applying Rang-Doppler Terrain Correction (Map Projection: WGS84(DD)) with output the latitude & longitude band on master_Stack_Deb, which you have mentioned before. And just copy the lon and lat bands to “geo” directory and renamed it. Is that correct? Should I apply terrain correction before export to StaMPS? I’m a little confused.
Many thanks!
So, I think I could get the lon and lat bands by applying Rang-Doppler Terrain Correction (Map Projection: WGS84(DD)) with output the latitude & longitude band on master_Stack_Deb, which you have mentioned before. And just copy the lon and lat bands to “geo” directory and renamed it. Is that correct?
Yes. I obtain lat and lon band when I apply TC correction.
Should I apply terrain correction before export to StaMPS?
Yes, you can do it. I export into StaMPS layers after TC because i must bypass shift problems.
OK! I will try this. And I just want to make sure about the input should be as below:
- stack_deb_TC.dim
- stack_deb_ifg_dinsar_TC.dim
And both of them need to choose “Latitude & Longitude” in output bands selection when applying TC?
But it only for the bypass shift problems, right?
For the topo phase removal error, you created interferograms for individual pairs manually to deal with that problem.
Is that the only way we could do with topo phase removal? Because for process testing, we could only use 4~5 images. For PSInSAR analysis, we need at least 12 images, that will be a little trouble.
Thank you very much!
But it only for the bypass shift problems, right?
Yes, it’s right.
And both of them need to choose “Latitude & Longitude” in output bands selection when applying TC?
Lat and lon band must be exported in binary file after StaMPS export, so it isn’t important the layer where they are. In my workflow, i export lat and lon band only in stack_deb_TC.dim.
Is that the only way we could do with topo phase removal?
In my opinion/experience now is only solution but it is very time expensive, also because if you want very accurate results, you must have at least 23-25 interferograms with little Bperp.
I hope that developers give us news as soon as possible. @ABraun, @marpet, @lveci, do you know something about toporemoval problems with stack or whom to ask?
1 Like
Hi @annamaria
Probably @junlu can say something about it. He’s the responsible developer fot the s1tbx, since @lveci left.
Maybe Andreas also knows something. But I have no clue. I’m not working with radar data at all.
2 Likes
I also have no experience with PSI in SNAP so far, sorry.
Thank you for your help!
hi @annamaria,
I have a small question.
Is it important the momento when I create a subset? I have prepared the stack of all the individual interferograms. At this point, and before to export to StaMPS, I would like to subset the image. Is it posible?
Thanks!
You can create subset at the end of your elaboration (before StaMPS export).
The important thing is that the size of subset must be the same of coregistated stack. So, In my opinion, you can use same geographic region both for interferogram stack and for coregistated stack.
@cfatras @FeiLiu @annamaria It seems that to avoid the NaN error in .base files it is necessary to select the SLC stack first and the interferogram stack second (I think that create stack tool does something strange with slave metadata).
Another question, which value for filter grid size parameter in stamps are you using to process Sentinel 1 data??
Hi annamaria,
Sorry for bothering you again and again.
I have tried to apply TC correction to get “stack_deb_TC” and “stack_deb_ifg_dinsar_TC” (this step took very long time about 4,200 mins, I only used 6 images).
And then put them into stamps export, I could only get two folders “dem” and “geo”, while losing “diff0” and “rslc” folders. If I used the products without TC correction into stamps export, I could get the correct four folders.
I don’t know is there any wrong in my processing? I haven’t tried to create interferograms for individual pairs and then create a stack, just apply TC on the deburst stack and dinsar results from katherine’s indications.
Thank you very very much!!
@alvegavaleri, @cfatras, @FeiLiu, @annamaria, and @junlu
Sorry for bothering all of you…
I have the same error as mentioned @alvegavaleri , in the output .base file appears NaN values.
I have checked the names of the coregistered images and ifg ones as mentioned @annamaria and have the same name.
I have redone the export step and the problem remains…
Have any of you got anything about it?
I am working with TC stacks, following the methodology described by @annamaria.
Thanks in advance!
Dear AranLarra,
I indeed had this issue only once, and by re-exporting the problem’s been solved. SNAP sometimes needs to insist a little bit!
I cannot help you more than that. Maybe a third exportaiton will be successful? Check the layer’s name maybe…
@AranLarra Have you tried reordering your stacks like a I said in my last message? (First include SLC stack, then IFG stack) Since I discovered this I haven’t had the issue again.
Dear All,
After running make install from src directory, I got this message, is that mean my installation is correct ??
After that when I am running mt_prep_gamma_snap then I am getting this error message,
Although, I have some output in the working directory,
I am really not getting what is wrong with me. Can anyone PLEASE help me.
Regards
Bayzidul