Resampling of S1 slave for offset (speckle) tracking?

Dear all,

  1. Is there a S1 offset (speckle or coherence) tracking sequence within SNAP?
  2. Otherwise, is it possible to coregister and resample a (full or subband) S1 slave image to a given master to apply offset (speckle) tracking afterwards? If so, what is the processing sequence?

Cheers, Jelte.

Offset tracking is planned to be developed in the next few months.

2 Likes

I saw that Offset tracking is now implemented in SNAP 3.0 - many thanks for that!
However, the help file in the documentation is still empty. Is there a publication or general algorithm it refers to?

I used offset tacking to know glacier velocity by using ALOS-1 satellite data. For that, I done coregistration and offset tracking as given in the “offse tracking tutorial”. When I opened velocity map it shows nothing (blank space). Please, let me know what is the solution for it…!

Hi Balu. ¿Where can I find the "“offse tracking tutorial”?

http://sentinel1.s3.amazonaws.com/docs/S1TBX%20Offset%20Tracking%20Tutorial.pdf

Dear all

I am using ALOS-1 data for offset tracking. I am getting different results when I am changing "grid azimuth spacing " and “range spacing” Processing parameters . Here any one know what are the best suitable values for the below parameters
1.Grid azimuth spacing
2. Grid Range spacing
3. Cross-correlation threshold

Thanks a lot Andreas. By the way, after coregistration I can’t find the residuals ¿Is there a way to get a list of the GCP offsets and residuals?

For getting GCP offsets, go to Generated velocity product —> Metadata —> Abstracted_Metadata —> Band_Velocity_Slave1_DATE —> Wrapdata

Yes, I would like to know the same, since I don´t understand what represents this grid exactly.
Hope sb knows, cheers!

grid space is the distance at which the result is calculated. If you select 1 km, for example, one value is retrieved per square kilometer. This is the main constraint which describes the results as all values in between are just interpolatet. So it has to be small enough to capture fine movement but too fine grid spaces cause longer processing time.

1 Like

Many thanks ABraun, then it is related to the velocity of movement also, is there a rule to select those grid values knowing the flow velocity?

I don’t know any - try-and-error :smiley: