This project uses different iterative methods for generalized linear inversion. The deconvolution and multiple suppression applications are also included in the package. The package can be used for any linearized inversion problem.
Different algorithms included in the package :
1)Steepest Descent with smoothness regularization
2)Steepest Descent with sparsirty regularization (Huber, Cauchy, Hybrid and Hoyer-squared norms)
3)Conjucate gradient with smoothness regularization
4)Conjucate gradient with sparsirty regularization (Iterative Re-wieghted Least squares Algorithm)
5)Fast Iterative Shrinkage Thresholding Algorithm
6)Alternating minimization Algorithm
We tested each algorithm on both synthetically generated seismic data and real data .You can get the code for each algorithm in the src directory .
The package is based on the work published in the Computers and Geosciences journal:
Naveen Gupta, Nasser Kazemi, PyInvGeo: An open-source Python package for regularized linear inversion in geophysics, Computers & Geosciences, Volume 202, 2025, 105948, https://doi.org/10.1016/j.cageo.2025.105948.
How to cite the paper:
@article{GUPTA2025105948,
title = {PyInvGeo: An open-source Python package for regularized linear inversion in geophysics},
journal = {Computers & Geosciences},
volume = {202},
pages = {105948},
year = {2025},
issn = {0098-3004},
doi = {https://doi.org/10.1016/j.cageo.2025.105948},
url = {https://www.sciencedirect.com/science/article/pii/S0098300425000986},
author = {Naveen Gupta and Nasser Kazemi}
}
We welcome comments and feedback.
-
The package is not heavily dependent on third-party solvers and libraries
-
All the codes are easily accessible
-
Development of new codes and ideas is easy
Data is saved in the data directory.
Seismic deconvolution and multiple supression for synthetic and real data
Results are autmatically stored at the folder results. It contain results with noisy data and noise-free data . It shows the deconvolution problem on 2D siesmic data set and multiple supression on synthetic and real data
It contains five Demo files. All the Demo files must be run in src directory .
1)Demo_decon.ipynb:-1D deconvolution problem has performed in this demo on synthetic data
link:-https://github.com/nasser00/Naveen-project/tree/main/src/Demo_decon.ipynb
2)Demo_syn_decon_2D.ipynb:-2D deconvolution problem has performed in this demo on synthetic data
link:-https://github.com/nasser00/Naveen-project/tree/main/src/Demo_syn_decon_2D.ipynb
3)Demo_decon_real_2D.ipynb:-2D deconvolution problem has performed in this demo on real data
link:-https://github.com/nasser00/Naveen-project/blob/main/Tutorials/Demo_decon_real_2D.ipynb
4)Demo_synthetic_radon.ipynb:- Multiple supression in synthetic data set
link:-https://github.com/nasser00/Naveen-project/blob/main/Tutorials/Demo_radon_syn.ipynb
5)Demo_real_radon.ipynb:-Multiple supression in real data set
link:-https://github.com/nasser00/Naveen-project/tree/main/src/Demo_real_radon.ipynb
matplotlib == 3.7.1
numpy == 1.25.0
scipy == 1.10.1
seaborn == 0.13.2