Normalize Incidence angle

Hello, dear all, how to normalize the incidence angle on S-1A image? For example, normalize the different incidence angle to 36?
Thank you very much.

  1. Convert to Beta0
  2. Run Radiometric Terrain Flattening

Please also see here:

4 Likes

Thank you, @ABraun, I’ve read the post you offered , we can normalize the incidence angle, but how to normalize the different incidence angle to 36 degrees? a specific number,

I have never heard of such a thing, sorry :smiley:
Why would someone do that?

Hello,

Unfortunately SNAP does not perform incidence angle normalization. This is an important step for SAR data acquired in wide swath mode such as SENTIENL-1.
Here is an interesting paper that explains how to perform incidence angle correction. https://www.degruyter.com/view/j/geo.2016.8.issue-1/geo-2016-0029/geo-2016-0029.xml

Thank you very much, @ABraun, In order to eliminate the error caused by the change of the local incidence angle ( Wagner, 1999a), In this forum, I learned a lot from your replies.

Thank you, @johngan, I’ll try and use the methods in the literature you provided!

I think SNAP actually does. Flattening to Gamma corrects radiometric distortions caused by topography. It is based on this work:
http://ieeexplore.ieee.org/document/5752845/
http://www.tandfonline.com/doi/abs/10.5589/m11-059?mobileUi=0&journalCode=ujrs20

1 Like

Hello mikellei.
I met this problem before. here is the link (Is there a fixed degree of angle for normalization in terrain correction?) I post several months ago, hope it might be helpful.

yes, gamma-naught is independent on local incidence angle.
However for some researchers like me or mikellei, the parameter we want is just normalized sigma-naught.

You can do an approximate incidence angle correction (i.e. normalization) for wide swath SAR images by assuming a simple relationship, like linear, between the incidence angle and backscattering coefficient. This relationship can be determined empirically using sample windows of sigma0 at different incidence angles. I have determined with my colleague incidence angle correction for the S-1 EW images over sea ice. The results can be also applied for other C-band images over sea ice, see
http://ieeexplore.ieee.org/document/8022896/

These incidence corrections are always approximate as a SAR image includes targets with different incidence angle vs. sigma0 relationships. But we have found that this approximate correction greatly improves SAR image classifications.

4 Likes

Well done! Thanks for your great work, mmakyen

The lookDirection of Sentinel-1 is Right only?

Yes, only right looking. You can refer to https://sentinel.esa.int/web/sentinel/missions/sentinel-1/instrument-payload

Thank you for your information, I’ve read it carefully these days, the method in this article can solve this problem, however, I am facing with a new problem: I can’t build a linear relationship between them,

Uploading…

Thank you very much, I can’t build a linear relationship between siggma0 and incidence angels inSNAP, I cant not upload a screenshot, the approache discussed in this paper is capable of solving this problem,

Thank you for your help, I’ll try it again,