search

slimgroup / JUDI.jl

Julia Devito inversion.
https://slimgroup.github.io/JUDI.jl
MIT License
96 stars 29 forks source link

Modeling LSRTM - Devito tutorial #180

Closed jddorellanao closed 1 year ago

jddorellanao commented 1 year ago

Hi,

I tried to replicate this Devito LSRTM in JUDI, so far I've done this. I'm having problems with the results, I'm 100% sure it's my coding for mixing these two references (1, 2) I am a newbie in JUDI.

The difference in amplitudes between the RTM image and the LSRTM image is huge. For the RTM I use:

imshow(diff(rtm.data, dims=2)', cmap="Greys", extent=[0, (n[1]-1)*d[1], (n[2]-1)*d[2], 0], vmin=-1e3, vmax=1e3)

Meanwhile for the LSRTM I use: 

imshow(diff(x.data, dims=2)', cmap="Greys", extent=[0, (n[1]-1)*d[1], (n[2]-1)*d[2], 0 ], , vmin=-0.05, vmax=0.05)

In the last figure I had to divide the RTM matrix by 1e4 to make the three lines somehow in the same range for the horizontal axis.

Thank you!

mloubout commented 1 year ago

That's normal and always the case. RTM doesn't have physical amplitudes so it is expected to have different scaling, LSRTM main goal is to go to physical amplitudes.