Giters

polimi-ispl / deep_prior_interpolation

Interpolation of seismic data through Deep Priors

Home Page:http://ispl.deib.polimi.it/research/geophysics/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

need suggestions for implimenting the deep prior int. tool on 2d seismic data

chiseo36 opened this issue · comments

commented

Hello, I am chiseo. Really, this is an amazing tool for interpolating the 3d seismic data. Could you please suggest me how to use this tool for 2d seismic data. In the 2d case i have the data dimension of mxn. However according to the instruction provided "If you have 2D native datasets, please add an extra axis". Thus i have added an extra axis by using the command inputdata[:, :, newaxis], in python and in this way i convert the 2d native data to 3d and now the data dimension is mxnx1. Though the data is now in 3d it gives something like "Invalid shape (n,)......So, i need some support from developers rectifying this small error.Thanks and looking forward to hearing from you.

commented

Hi @chiseo36,

Thanks for trying this method. Please, refer to this issue and let me know if the problem persists.

commented

Many Thanks @fpicetti for your hard work. It's running fine now. For one thing i need more suggestions, if i use this code proof_of_concept2D for my research publication i think it is valid as it is already published in a reputed journal and it can be cited, however antialiasing_lines.ipynb method is good but it is not included in any reputed journal, nevertheless it's implication has been shown here https://arxiv.org/abs/2101.11361. But many researcher believe that arxiv.org papers cannot be cited. So can you please calrify it if i use antialiasing_lines.ipynb how can i cite it.

commented

Hi @chiseo36, thanks for asking :)
That paper has been accepted at IEEE ICIP 2021, that will take place in September. I'm going to add that publication to the list on the README file, once the conference is done.

We've done a lot of research on this deep prior interpolation, still there are many ideas to work out. We will be very interested to follow yours.
Also, feel free to contribute to this repo, by suggesting enhancements and opening pull requests.

commented

Thanks @fpicetti. I am implementing antialiasing_lines.ipynb on both synthetic and real field data set. For synthetic data i did it by masking the data and moreover it has also target data, however while implementing it to the field seismic data I am facing problem.

My field input data has already gaps(looks like masked data) , i want to fill the gaps using this mentioned method. My query is does it require to mask the data at this time also, if so what would be the missing percentage for masking(does it 0.001) and what should be the target here (which is a filled data as in synthetic case)? (but i don't have filled data, which i want to get ). Additionally, i tried with setting missing percent to 0.001 and took gaped data as input at this time it does not fill anything, it shows PCORR=100% SNR=+inf. Can you please suggest a solution regarding this.Thanks in advance and looking forward to hearing from you.