Dear colleagues
How can remove the light white color in SAR Sentinel-1 sigma VV backscattering images . It confused me that I ask is it a subsurface buried feature or not?
the preprocessing of sentinel-1 for the GRD data adopted (Apply orbit, Thermal noise remover, Remove GRD Border Noise, speckle filtering, Terrain Correction) and stay same the feature color in output VV sigma bands and don’t remove.
any clarification is welcomed.
is it only in this one image or persistent throughout all scenes?
Looks a bit like Radio Frequency Interference.
Please check and compare here:
- Stripes near Copenhagen
- Sentinel-1 noise - can it be removed?
- Noise objects in Sentinel-1 GRD data
- Artifacts in Sentinel-1 GRD data. How to detect/filter?
Where is your scene located?
Dear Mr. ABraun
Thanks for clarification the noise just in one image in VV polarization, only in this place the image located in Iraq… Sometimes I saw features with white color in VV Polarization and with shady pink/rose color in RGB color Image VV/VH. Are these features considered RFI?.
@Meelad_16777 My understanding is that this is indeed RFI. It makes sense that it is pink in the RGB image, as it is a it appearing in the Red and Blue color channels.
Out of blatant curiosity, what image is this? Sometimes I like to go and check out RF interference and try to determine the source.
I don’t want to speculate too much, but this is a very interesting article on the effect of artillery on Sentinel-1 images:
2 Likes
Dear @ABraun and @BarrowWight
Thanks for your clarification, I feel disappointment when I discovered that these stripes as noise and not just subsurface buried structures so you see me doesn’t reacting with your post. Indeed it just RFI… These noises don’t appear in all SAR Sentinel Images over the year… Great Thanks for you…
@Meelad_16777 Agreed. I recently was talking to a professor I know, and he said that SAR satellites are not really capable of ground penetration more than a few millimeters. You would need lower frequencies to beyond that apparently, which is a pity as I first got interested in the Sentinel-1 data as a way to identify buried structures for archeology.
@ABraun that’s cool
1 Like
@BarrowWight
I’m agree with you, the penetration of Sentinel-1 of C Band are very more difficult to infer it require some of conditions and parameters ((bare soil, low Incidence angle, low frequency)) as so you need big data for temporal time series of Sentinel-1 for capture the variability of archaeological marks, which it in the best condition some of centimeters. Recently I’m work on the ALOSPALSAR L band with several meters as mentioned in some of paper.
Great thanks for your interest.
Dear @ABraun
Thank you for sharing this presentation, I don’t find any source of Patriot Batteries near my area and as see before the stripes just in on place. Could you see the Sentinel-1 SAR imagery below, because I hope that some of it not RFI.
Bath 6 Frame 486
S1B_IW_GRDH_1SDV_20190131T025354_20190131T025428_014732_01B780_8062
S1B_IW_GRDH_1SDV_20190107T025355_20190107T025428_014382_01AC3E_0939
S1B_IW_GRDH_1SDV_20181226T025355_20181226T025429_014207_01A685_20B3
S1B_IW_GRDH_1SDV_20180617T025351_20180617T025425_011407_014F37_746A
S1B_IW_GRDH_1SDV_20180418T025348_20180418T025421_010532_013348_4C76
S1B_IW_GRDH_1SDV_20180430T025349_20180430T025422_010707_0138E3_7059
S1B_IW_GRDH_1SDV_20180418T025348_20180418T025421_010532_013348_4C76
Note these Sentinel data available in ASF website and can show these noises in preview window.
Wait your Opinion.
Quickly skimming through the images, I would say that all of those perfect squares that are a stuttering (for a lack of better words), are indeed probably noise.
Examples of what I mean:
Dear @BarrowWight
What about this scene
(S1B_IW_GRDH_1SDV_20180430T025349_20180430T025422_010707_0138E3_7059).
The scene represents bare soil.
Thanks for interest.
@Meelad I would guess that is also noise. The fact that it is a perfect rectangular shape and how it seems to not be affected by anything “under” it are the takeaways I have.
2 Likes
We are seeing lots of RFIs in Denmark.Looking at a 2 year time series they change size and position. Sometimes they are very faint, what is very tricky.
Btw. there is a Google Earth kmz which shows location of ground based radars at SAM sites (not up to date, of course) which may interfere with Sentinel-1 C band. https://www.gearthblog.com/wp-content/uploads/2016/04/SAM-sites.kmz
These authors have also taken a look at them and are still working on a solution on detecting them.
2 Likes
SAR sensors are good for ground penetration well past a few mm, depending on the conditions. There are many articles on finding subsurface features in the deserts. Using spaceborne SAR for archaeology is a fairly new discipline but it is becoming more popular with more research being available every day.
The following are good examples:
https://ieeexplore.ieee.org/abstract/document/4072513
3 Likes
Dear @fwiig
The great challenge of these researches is identification of the depth of the buried archaeological structures or features. The question is “” What is the depth of the archaeological features that can SAR Sentinel-1 enabled to discover. as for as The ALOS PALSAR with L-Band (23m) the depth that capable to discover is 2 m in desert sites, where the Sentinel-1 SAR C-Band (5.7) depth calculated to (0.2-0.5).
For example
1- https://www.mdpi.com/2072-4292/12/16/2611
2- https://www.mdpi.com/2072-4292/9/6/638/pdf
Yes that is the big challenge. And i am doing my PhD on this. And Interestingly have had more success with X-Band than the other wavelengths although this is also due to the finer resolution at that particular site.
Dear fwiig
Can you please inform me of the steps how overcomes these challenges by guiding me for your work (Paper, Thesis, Presentations, any published pinpoint).
I welcome any explanation
If I understand your post, do you mean the X band belong to SRTM DEM Satellite images.? or another Active Satellite.
Regard
Hello,
I have used the TanDEM-X/TerraSAR-X X-band data with success. You can see my article in this chain above: Mapping a subsurface water channel with X-band and C-band SAR…
Good luck