Cross Correlation/ Terrain Correction: not enough valid GCP's for the warp

Microwave Toolbox
1

Hello SNAP forum,

I am attempting to orthorectify and terrain correct S1 IW imagery using the attached batch processing graph (step1), before i test some speckle filters (step 2). As displayed in the command line interface below, i receive an error stating there are" not enough valid GCP’s for the warp" during the cross correlation process. The result is a blank screen in SNAP when displayed for all bands (except for the local and projected incidence angle). I have changed the number of GCP’s from a range of 200 up to 2000 points, but i still receive the same error. If you examine the xml graph attached you will see I am using an external DEM and a subset in this process as well; when i do not subset, the SAR simulation/ cross correlation works with no GCP error regardless of the GCP setting.

Is the source of this error another parameter in the Cross correlation setting ? Or, is this a subsetting compatibility issue with my current process? I would prefer to keep the subset node in place to reduce processing time.

Thank you for any suggestions & advice

STEP1_S1.xml (6.5 KB)

NotEnoughGCPs.PNG1282×307 8.49 KB

2

Hi @Benk

Have you find a solution for this problem?
I have been running into a similar issue for some of my images.

Thanks,
Dieuwertje

3

For some cases, the topographic pattern is not suitable for coregistration with a simulated SAR image (generated from the DEM).

Is Range Doppler terrain correction an option instead?
Is the external DEM projected in WGS84?

4

Thanks for your reply @ABraun!

The external DEM is in WGS84 zone 33N, which I do set in . I have used images from the same region and track but different date and for most it works, but for some I get this error.

As I do want a layover and shadow mask, so that is why I have been using the SAR Simulation Terrain Correction.

5

it is either WGS84 (latitude/longitude) or it is UTM zone 33N. Please make sure it is projected to geographic coordinates (WGS84, EPSG:4326), otherwise SNAP cannot read or allocate it. This maybe explains why it couldn’t be used in the

grafik