I would like to know where the issue lies. I have uploaded my .in file and B-SCAN. Could you please assist me?Why is there a blurred concave region in the B-SCAN

[王晨]

[Craig Warren]

Can you post a screenshot of the geometry view please.

Craig

On Thursday 8 February 2024 at 00:35:14 UTC chen...@gmail.com wrote:

[liu ft]

I also encountered this situation when drawing cross-polarized images, how did you solve it later?

在2024年2月8日星期四 UTC+8 08:35:14<chen...@gmail.com> 写道：

[Craig Warren]

Please post an image showing your geometry.

Craig

[liu ft]

Thank you for your reply!
This is my .in file

When I drew the Bscan for Ex and Ey directions, the following results appeared:

Ex：

Ey：

I have two questions:
① : Why is the hyperbola incomplete in cross-polarized Bscan images
② : Why is the cross-polarized signal stronger than the positively polarized one, even after removing the direct wave

Looking forward to your reply, thanks!

Liu ft

[Craig Warren]

Please post an image of your geometry (mesh) generated from your input file, i.e. by using a geometry_view in your input file.

Craig

[liu ft]

Thank you for your reply. This is my model, in which the transceiver antenna is too small, I circled it in red.

[Antonis Giannopoulos]

You are looking at the Ey component which is a vertical component for your geometry. This is not the cross-polarised response that the Ez will give you as a good estimate (there is a small offset in the components and it is not exactly cross-pol) of GPR cross-polarised response. 

The vertical component above a metal target approaches a minimum directly over it. It looks asymmetric because it is the field received above the target irrespectively of the position of the transmitter. In your Ex scan the apex of the hyperbola is at the mid-point of your Tx and Rx as it should. However, the vertical component is at a minimum directly over the target. That means that it should appear before the apex and when the Rx itself is above the target and not when the mid-point of the Tx and Rx is over it.

I have done a similar analysis in my PhD thesis very very long time ago looking in the response of the magnetic fields but the concepts are the same.

If you are interested in the cross-polarised response  that some GPRs provide  then you are looking at the wrong component. In this case you need the Ez one. Your model is otherwise OK.

Hope this helps

Antonis

[liu ft]

Thank you for your reply!

I have another question. When I set the transmitted wave in the x-direction and plot the B-scan of the Ez component, its amplitude appears very small (compared to the previous two cases), even after removing the direct wave using the mean value method. Additionally, the B-sscan seems to have slight gaps.

Looking forward to your reply.

Liuft