EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An unsupervised inversion framework in the frequency domain using a Wasserstein generative adversarial network

Cong Wang, Xingguo Huang, S Mostafa Mousavi, Yue Li

Geophysical Journal International · 10.06.2026 · 246 (2)

DOI: 10.1093/gji/ggag192 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Geophysical Journal International, online erschienen am 15.05.2026. Verfasst von Cong Wang und 3 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, National Science and Technology Major Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reconstructing subsurface structures with high resolution is one of the main goals and potentials of full waveform inversion (FWI). However, FWI is a highly nonlinear and ill-posed problem. Conventional physics-based FWI methods, which rely on gradient-based optimization to minimize the difference between observed and synthetic data face cycle-skipping challenge. Although numerous deep-learning inversion approaches have shown promise, they typically focus on latent representations of time-domain seismic data. This often causes an unstable inversion process due to waveform mismatches. To overcome these limitations, we introduce an unsupervised seismic inversion framework that integrates physics-based forward modelling with adversarial learning of the frequency-domain data based on Wasserstein generative adversarial network with gradient penalty (FFT-InversionGAN). Fast Fourier transformer (FFT) is employed to transfer the time and phase information of time-domain seismic data into the spectrum and amplitude distributions to modify the feature space and sensitivity of the adversarial loss to different types of mismatches. By leveraging Wasserstein distance constraints, this method can naturally operate on the spectral distributions of seismic data. Compared with L2 norm, Wasserstein distance is far less sensitive to the linear variations in the phase spectrum. And our proposed method eliminates the need for network pre-training while improving stability and flexibility. FFT-InversionGAN demonstrates enhanced accuracy and resilience in numerical experiments on noise-free, noisy and missing low-frequency benchmarks. This was observed when applied to the Marmousi and overthrust models, where it consistently outperformed conventional FWI and FWI based on generative adversarial network (FWIGAN). These findings highlight that FFT-InversionGAN has superior inversion effectiveness.

Förderung

  • National Natural Science Foundation of China
  • National Science and Technology Major Project

Zitieren

APA

Wang, C., Huang, X., Mousavi, S. M., & Li, Y. (2026). An unsupervised inversion framework in the frequency domain using a Wasserstein generative adversarial network. Geophysical Journal International, 246(2). https://doi.org/10.1093/gji/ggag192

Vancouver

Wang C, Huang X, Mousavi SM, Li Y. An unsupervised inversion framework in the frequency domain using a Wasserstein generative adversarial network. Geophysical Journal International. 2026;246(2). doi:10.1093/gji/ggag192

BibTeX
@article{Wang2026,
  author = {Wang, Cong and Huang, Xingguo and Mousavi, S Mostafa and Li, Yue},
  title = {An unsupervised inversion framework in the frequency domain using a Wasserstein generative adversarial network},
  journal = {Geophysical Journal International},
  year = {2026},
  volume = {246},
  number = {2},
  doi = {10.1093/gji/ggag192}
}
Kostenlos starten →