S1 Backgeocoding Error

Microwave Toolbox Problem Reports
#1

Greetings!

I have been trying to co-register two tiles of S1-TOPS data using S1 backgeocoding tool in SNAP 8.0, the processing gets completed smoothly without any error. However, when I open the slave image it has NaN values.
I tried using SNAP 7.0 also but the same result.
I tried changing master and slave bands but always the slave image shows NaN values with min and max values of 0.0 and 4.9E-324 respectively.

Any help would be appreciated @ABraun @jun_lu @mengdahl @marpet

Thanks and Regards

#2

An empty band after BackGeocoding means that the coregistration failed. Did you install all updates?

can you please share the product IDs?

#3

Sorry for delayed reply,

Yes I installed all the updates and tried again but the result is same

The Product ids are:
S1A_IW_SLC__1SDV_20200813T005905_20200813T005935_033881_03EDEC_0DDA
S1A_IW_SLC__1SDV_20200825T005906_20200825T005936_034056_03F420_D50C

#4

have you reduced the subswaths during TOPS Split to the same area?

#5

Yes, 21001 * 5972 pixels in both products after doing TOPS split

#6

can you plase share a screenshot of the two input products for BackGeocoding and which preprocessing you applied so far?

#7

image
image683×534 16.7 KB

S1 TOPS Split ----> Apply Orbit File -----> S1 Back Geocoding

#8

did you open them and visually check for valid data?

#9

Master Image is fine and pixel values are valid

Slave image shows NaN pixel vlaues

#10

Sorry, I meant before the coregistration

#11

Yes, the data is valid!

I checked after subsetting to small area also, but the same results.

#12

at which stage did you apply the subset? Actually, you shouldn’t apply any spatial subset before the coregistration (and debursting) of the images when working with SLC data

#13

I tried coregistration by subsetting the data and without subsetting the data also.

#14

The processing steps employed are as:

  1. SLC image —> S1 TOPS split —> Apply Orbit file —> S1 Backgeocoding

  2. SLC image —> S1 TOPS split —> Apply Orbit file —> Spatial subset —>S1 Backgeocoding

both pathways yield the same result

#15

this one should work actually, but I’m out of ideas why you receive a partially blank output. Have you tried using another DEM for BackGeocoding? E.g. SRTM 1Sec instead of 3Sec

#16

Yes, i am using SRTM 1 sec

and it is not first time , I have tried corregistration on two SLC tiles for the year 2017 over upper himalayas, there also the error occured. Fore some image pairs it is working, for others it doesnot.

#17

which sub-swath and which bursts did you use in TOPS Split?

#18

Subswath IW3
Burst 4 to 7

#19

I have tried your data with the mentioned steps and configurations and did not receive any error.

grafik
grafik1247×948 632 KB

1 Like
#20

can you send me the interferogram file, please?

Does it have to do with RAM overload, my machine has 16 GB RAM?