In this branch, we provide a new feature, jax-based Seistorch. The jax-based Seistorch is designed to provide a more efficient (10x up) and flexible way to solve wave equations and perform seismic inversion tasks.
The jax-based Seistorch is still under development, and we welcome any feedback or contributions.
| Inversion Tests | Status |
|---|---|
| Acoustic | Passed |
| Elastic | Passed |
| Others | Not test yet |
I reproduced the results of the following papers using Seistorch and some stand alone codes. If you are interested in these topics, please refer to the following links:
| Traditional | Codes | Related Papers | Notes |
|---|---|---|---|
| Simulations | click | Wang et al., 2023 | Seistorch |
| Finite difference method | click | - | Acoustic |
| Pseudospectral method | click | Kosloff & Baysal | Acoustic |
| Traditional | Codes | Related Papers | Notes | Support by |
|---|---|---|---|---|
| FWI by Pytorch | click | - | Stand alone | Pytorch |
| FWI by Jax | click | - | Stand alone | Jax |
| Acoustic FWI | torch,jax | - | Seistorch | Pytorch/Jax |
| Elastic FWI | torchjax | - | Seistorch | Pytorch/Jax |
| Regularization-based FWI | click |
| Traditional | Codes | Related Papers | Notes | Pytorch | Jax |
|---|---|---|---|---|---|
| Acoustic LSRTM | click | Dai et al., 2010 | Seistorch | ✓ | ✓ |
| Elastic LSRTM | click | Feng & Schuster, 2017 | Seistorch | ✓ | x |
| VTI/TTI LSRTM | click | - | Seistorch | ✓ | ✓ |
| Joint FWI&LSRTM | click | Wu et al., 2024 | Seistorch | ✓ | x |
| Regularization-based LSRTM | click | ✓ | x |
| FWI+NeuralNetworks | Codes | Related Papers | Notes |
|---|---|---|---|
| PINN | click | Majid et al., 2022 | Stand alone |
| Implicit FWI | click (Pytorch) click (Jax) |
Sun et al., 2023 | Stand alone Model Reparameterization(Acoustic) |
| Physics-guided NN FWI | click | Dhara & Sen, 2022 | Stand alone Model Reparameterization(Acoustic) |
| Elastic parameters crosstalk | click | - | Stand alone Model Reparameterization(Elastic) |
| Siamese FWI | click | Omar et al., 2024 | Stand alone |
| Elastic parameters crosstalk | click | Dhara & Sen | Stand alone Model Reparameterization(Elastic) |
| Misfits | Examples | Related Papers | Notes | Pytorch | Jax |
|---|---|---|---|---|---|
| Optimal Transport | click | Yang & Ma, 2023 Yang & Enguist |
- | ✓ | x |
| Envelope | click | Chi et al., 2014 Wu et al., 2014 |
- | ✓ | ✓ |
| Traveltime | click | Wang et al., 2024 | Differentiable | ✓ | x |
| Cosine Similarity | click | Choi & Alkhalifah, 2012 Liu et al., 2016 |
Global correlation Normalized zero-lag cross-correlation |
✓ | x |
| L1 | click | ✓ | ✓ | ||
| L2 | ✓ | ✓ | |||
| Local coherence | click | Yu et al., 2023 | - | ✓ | x |
| Instantaneous Phase | click | Bozdag et al., 2011 Yuan et al., 2020 |
- | ✓ | x |
| Weighted loss | click | Song et al., 2023 | ✓ | x | |
| Envelope Cosine Similarity | * | Oh and Alkhalifah, 2018 | Envelope-based Global Correlation Norm | ✓ | x |
| Soft Dynamic Time warpping | click | Maghoumi, 2020 Maghoumi et al., 2020 |
✓ | x |
| Type | New | Old | Notes |
|---|---|---|---|
| Backend | Jax | Pytorch | 10x up |
| EQUATIONS | USAGE | REFERENCES | EQUATION CODES | Pytorch | Jax |
|---|---|---|---|---|---|
| Scalar Acoustic (2nd) | FWI | * | PML version HABC version |
✓ | ✓ |
| Scalar Acoustic (2nd) | LSRTM | Dai et al., 2010 | click | ✓ | ✓ |
| Acoustic (1st) | FWI | * | click | ✓ | x |
| Variable Density (2nd) | FWI | Whitmore et al., 2020 | click | ✓ | x |
| Joint FWI & LSRTM | FWI+LSRTM | Wu et al., 2024 | click | ✓ | x |
| qP TTI (2nd) | FWI/LSRTM | Liang et al., 2024 | fwi click lsrtm click |
✓ | ✓ |
| qP VTI (2nd) | FWI/LSRTM | Liang et al., 2024 | fwi click lsrtm click |
✓ | ✓ |
| ViscoAcoustic (2nd) | FWI | Li et al., 2016 | click | ✓ | x |
| VTI (2nd) | FWI | Zhou et al., 2006 | click | ✓ | x |
| Elastic (1st) | FWI | Virieux, 1986 | click | ✓ | ✓ |
| Elastic (1st) | LSRTM | Feng & Schuster, 2017 | click | ✓ | ✓ |
| TTI-Elastic (1st) | FWI | * | click | ✓ | x |
| Acoustic-Elastic coupled (1st) | FWI | Yu et al., 2016 | click | ✓ | x |
| Velocity-Dilatation-Rotation (1st) | FWI | Tang et al., 2016 | click | ✓ | x |
Note: 2nd means displacement equations, 1st means velocity-stress equations.
- Make all x codes to ✓ codes.
If you find this work useful for your research, please consider citing our paper Memory Optimization in RNN-based Full Waveform Inversion using Boundary Saving Wavefield Reconstruction:
@ARTICLE{10256076,
author={Wang, Shaowen and Jiang, Yong and Song, Peng and Tan, Jun and Liu, Zhaolun and He, Bingshou},
journal={IEEE Transactions on Geoscience and Remote Sensing},
title={Memory Optimization in RNN-based Full Waveform Inversion using Boundary Saving Wavefield Reconstruction},
year={2023},
volume={61},
number={},
pages={1-1},
doi={10.1109/TGRS.2023.3317529}}