Stack with Pins (S1) - NaN for some Pins

Microwave Toolbox
#1

Hi Guys,
i have some S1-GRD-Images that I processed the following way:

Read -> Thermal Noise Removal -> Apply Orbit file -> Subset -> Calibration -> Speckle-Filtering (Lee 5x5) -> Terrain-Correction (Range-Doppler).

Afterwards I used Coregistration to create a stack of several images and placed pins on one band. I transferred those to all other bands and extracted the values for my “bands” (VV and VH for different times).

For some pins, in some images, I get NaN Values extracted at this pins, which does not make a lot of sense to me.

Would be nice if you could help me with this, I really need those values :smiley:

With best regards,
Felix

#2

Similar issue is solved in here,

Source of the post

#3

Thank you! I somehow did nor search for NaN in the search-function :smiley: Thank you so much!

#4

Dear falahfakhri,
You happen to know why this is only apperaing on some images?

If I did the coregistration, shouldn’t there be “lake” on all images for some pins? Its not like the lake has high dynamics in extent, so the shoreline is pretty much the same on all images.

I asked the same question in the other thread, but maybe you know it and see it faster in here.

Additionally, I want to know which kind of coregistration is the best. I used the standard one but I read that the DEM-assisted one is better and more precise. Are there known issues if the images are not 100% overlapping? like some parts of the lakes are missing in some images?
Best,
Felix

#5

Do you mean without checking the box of (mask out area without elevation)?

Does your study area only the lake, (water body)?

Would you please to share a screenshot to understand more,

Is your study area terrain or water?

#6

Yes, I initially didnt do this, since I did not know. (It is not relevant for my current work, just interested in genereal, but maybe it has something to do with this Create Stack with different image parts).

I stacked the images (~20) and set the pins (~15) inside the lake boundaries(I am only interested in the lake values), so far so good. Then I extracted the values.

For some pins - in some images (not all) - i got NaN, for some I got values.

The part about the shoreline was just to clarify that all the pins will be inside the lake on all images (if they were stacked right, which I somehow do not think after the thread above).

I am interested in lake-values only :slight_smile: