error SBAS stamps(1,1)

[Antonella Belmonte]

Dear all, 
I've encountered a problem with Smallbaseline processing for TSX images: 
after succesful finishing PS processing I run in INSAR_masterdate 
matlab 
>>sb_find(0.5,150,1700) 

            ( all folks ) 
make_small_baselines 
cd SMALLBASELINES/ 
mt_prep 0.6 10 10 100 100 

matlab

>>stamps(1,1)

########################################

############ StaMPS Step 1 #############

########################################

SETPARM: heading = 191.0367

SETPARM: lambda = 0.031

Attempted to access llh(2,:); index out of bounds because size(llh)=[0,0].

Error in llh2local (line 39)

   z=llh(2,:)~=0;

Error in sb_load_initial (line 142)

xy=llh2local(lonlat',ll0)*1000;

why?

Best Regards

Antonella

[David Bekaert]

Hi,

Did you by chance forget to run the geocoding step?

Cheers,

David

--
You received this message because you are subscribed to the Google Groups "MAINSAR" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mainsar+u...@googlegroups.com.
To post to this group, send email to mai...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Antonella Belmonte]

Thanks for your answer,

but I have run the geocoding step during the pre-pocessing before the PS-STAMPS algorithm inside only one of the slave directories.  Now in SBAS,  is it necessary  to geocode an another time?

I have read on group that if  I decrease the AF_BLOCKSIZE in filtazi.dorisin probably the error disappears.

I do it with the size 1024. I run make_small_baselines to obtain the file ‘AF_BLOCKSIZE’, then I stop, I change the dimension and rerun make_small_baselines deleting the first interferogram subdirectory, but nothing change. Where do I make a mistake?  

[David Bekaert]

Hi,

Its enough to have step_geo ran during the PS processing.

I also have seen this error when mt_prep fails.

You could check if the selected candidate data in each patch after mt_prep is reasonable. 

Also if there are files missing as when the patches are too large they can contain too much data (> larger than 2 GB) matlab will fail to save it.

You should have seen a message of that during mt_prep.

Also are there any patches that cover water?

Cheers,

David

 

[Antonella Belmonte]

Yes, there are some patches that cover water.

[Antonella Belmonte]

When I run SBAS  all the pscands files in every Patch folder  are 0 size.

If now I delete the water-Patch during SBAS  in which all pscands files are size 0, can I resolve the problem?

I have not problem during the processing PS-STAMPS using all PATCHES.

Cheers

Antonella

[Karsten Spaans]

Hi Antonella,

You can modify the patch.list file in the INSAR_XXXXXX directory by removing the 0 size patches. StaMPS will then skip these patches.

Cheers,
Karsten

Op 10-12-2014 11:11, Antonella Belmonte schreef:

[Alberto Refice]

Hi all,

we have found a possible wrong handling of the buffer size in the c code
called by the mt_prep script.

The error Antonella describes appears when processing through SBAS large
stacks of TerraSAR-X images. In such cases, the script that analyzes the
data to extract PS candidates reads in data through a buffered read.

The buffer size is fixed in the code to 100 image lines, but apparently,
when the total size of the read buffer (buffer_lines * image_width *
number_of_interferograms) exceeds 2 GB, there is an overflow, which does
not cause a segmentation fault, but simply causes the output to be null.
As a consequence, the files containing ps candidates in the patch dirs
are empty, and this sends astray all the subsequent processing.

Maybe this is connected to some issue with the gcc version (ours is
4.4.5), anyway decreasing the buffer size (to e.g. 10 lines) solves the
problem.

Cheers,

Alberto

> --
> You received this message because you are subscribed to the Google
> Groups "MAINSAR" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to mainsar+u...@googlegroups.com

> <mailto:mainsar+u...@googlegroups.com>.

> To post to this group, send email to mai...@googlegroups.com

> <mailto:mai...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout.

--

Alberto Refice
Istituto di Studi sui Sistemi Intelligenti per l'Automazione -
Consiglio Nazionale delle Ricerche
Via Amendola 122/d -- 70126 Bari (Italy)
TEL: +39 080 5929432 FAX: +39 080 5929460
EMAIL: ref...@ba.issia.cnr.it - WEB: http://www.issia.cnr.it/

[陈宇]

Hi Alberto,

I got a similar problem when I run mt_prep. the output files in each patch folder are all zeros.

 I did not get any messages during the processing. but it is obvious that there must be something 

went wrong because it took just several seconds when runing "mt_extract_cands". my image width is 7801,

 I have 324 interferograms in total. I am wondering if it is due to the reason that you mentioned? could you 

tell me which file I should edit  to change the buffer size? 

Thank you very much!

Yu

[Alberto Refice]

Hi Yu,

I attach the patched version of the selsbc_patch_new.c source file. See
if it works.

Thanks to Francesco Lovergine for the correction!

Regards,

Alberto

> > an email to mainsar+u...@googlegroups.com <javascript:>
> > <mailto:mainsar+u...@googlegroups.com <javascript:>>.

> > To post to this group, send email to mai...@googlegroups.com

> <javascript:>
> > <mailto:mai...@googlegroups.com <javascript:>>.

> > For more options, visit https://groups.google.com/d/optout

> <https://groups.google.com/d/optout>.

>
> --
> Alberto Refice
> Istituto di Studi sui Sistemi Intelligenti per l'Automazione -
> Consiglio Nazionale delle Ricerche
> Via Amendola 122/d -- 70126 Bari (Italy)
> TEL: +39 080 5929432 FAX: +39 080 5929460

> EMAIL: ref...@ba.issia.cnr.it <javascript:> - WEB:
> http://www.issia.cnr.it/

[Alberto Refice]

Another possible improvement is the attached diff file to implement
64-bit size_t allocation... this would allow using arbitrary buffer sizes...

Hope this helps,

Alberto

[陈宇]

Hi Alberto,

Thanks a lot, it works very well. I really appreciate it!

cheers

Yu

[Mike Johnson]

Hi Antonella,

     I'm new to StaMPS,can you kindly provide some detailed tutorials about StaMPS to me(except for the StaMPS manual) ? Thank you very much!

在 2014年12月5日星期五 UTC+8下午10:56:01，Antobel写道：

[Roqaye]

Hi Alberto

I have the same problem, but don’t know how to use the ''selsbc_patch_new.c'' script, which you attached.
I would appreciate it if you can help me.

Kind regards
Roghayeh

[Jerlyn Swiatlowski]

Hi Karsten, 

  In your reply when you said that you can modify the patch.list file to not include the directories with 0 patch sizes, doesn't this take out a section of the over all area you are trying to process? Or is it different from what is happening in the PS processing? Why can we just remove it? 

Thank you, 

Jerlyn