Skip to content

Plug and Play Post-Stack Seismic Inversion with CNN-based Denoisers

Notifications You must be signed in to change notification settings

DIG-Kaust/Seismic_PnP

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

52 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

LOGO

Reproducible material for Plug and Play Post-Stack Seismic Inversion with CNN-based Denoisers
Romero J., Corrales M., Luiken N., and Ravasi M.

submitted to Second EAGE Subsurface Intelligence Workshop, 28-31 October 2022, Manama, Bahrain

Project structure

This repository is organized as follows:

  • 📂 data: Marmousi impedance synthetic model.
  • 📂 models: folder containing pre-trained models (DnCNN and DRUnet).
  • 📂 notebooks: jupyter notebook reproducing the experiments in the paper.
  • 📂 pnpseismic: package of the project (Deep denoisers architectures and PnP framework).

Notebooks

The following notebooks are provided:

  • 📙 PnP_PD_Post-Stack_Seismic_Inversion_marmousi.ipynb: notebook performing the comparison between model-based regularization and Plug and Play.

Getting started 👾 🤖

To ensure reproducibility of the results, we suggest using the environment.yml file when creating an environment.

Simply run:

./install_env.sh

It will take some time, if at the end you see the word Done! on your terminal you are ready to go. After that you can simply install your package:

pip install .

or in developer mode:

pip install -e .

Remember to always activate the environment by typing:

conda activate pnpseismic

Disclaimer: For computer time, this research used the resources of the Supercomputing Laboratory at KAUST in Thuwal, Saudi Arabia. All experiments have been carried on a Intel(R) Xeon(R) Platinum 8260 CPU @ 2.40GHz equipped with a single NVIDIA TESLA V100. Different environment configurations may be required for different combinations of workstation and GPU.

Pre-trained Models

For more details of the Pre-trained Deep Denoisers used in this study, please refer to the following repositories: https://github.com/cszn/DnCNN and https://github.com/cszn/DPIR.

Cite us

@article{eage:/content/papers/10.3997/2214-4609.2022616015,
   author = "Romero, J. and Corrales, M. and Luiken, N. and Ravasi, M.",
   title = "Plug and Play Post-Stack Seismic Inversion with CNN-Based Denoisers", 
   journal= "",
   year = "2022",
   volume = "2022",
   number = "1",
   pages = "1-5",
   doi = "https://doi.org/10.3997/2214-4609.2022616015",
   url = "https://www.earthdoc.org/content/papers/10.3997/2214-4609.2022616015",
   publisher = "European Association of Geoscientists & Engineers",
   issn = "2214-4609",
   type = "",
  }

About

Plug and Play Post-Stack Seismic Inversion with CNN-based Denoisers

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages