Feature-based Inversion of 2.5D Controlled Source Electromagnetic Data using Generative Priors
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908748322177024 |
|---|---|
| author | Zhou, Hongyu Sun, Haoran Guo, Rui Li, Maokun Yang, Fan Xu, Shenheng |
| author_facet | Zhou, Hongyu Sun, Haoran Guo, Rui Li, Maokun Yang, Fan Xu, Shenheng |
| contents | In this study, we investigate feature-based 2.5D controlled source marine electromagnetic (mCSEM) data inversion using generative priors. Two-and-half dimensional modeling using finite difference method (FDM) is adopted to compute the response of horizontal electric dipole (HED) excitation. Rather than using a neural network to approximate the entire inverse mapping in a black-box manner, we adopt a plug-andplay strategy in which a variational autoencoder (VAE) is used solely to learn prior information on conductivity distributions. During the inversion process, the conductivity model is iteratively updated using the Gauss Newton method, while the model space is constrained by projections onto the learned VAE decoder. This framework preserves explicit control over data misfit and enables flexible adaptation to different survey configurations. Numerical and field experiments demonstrate that the proposed approach effectively incorporates prior information, improves reconstruction accuracy, and exhibits good generalization performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_02145 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Feature-based Inversion of 2.5D Controlled Source Electromagnetic Data using Generative Priors Zhou, Hongyu Sun, Haoran Guo, Rui Li, Maokun Yang, Fan Xu, Shenheng Geophysics Machine Learning In this study, we investigate feature-based 2.5D controlled source marine electromagnetic (mCSEM) data inversion using generative priors. Two-and-half dimensional modeling using finite difference method (FDM) is adopted to compute the response of horizontal electric dipole (HED) excitation. Rather than using a neural network to approximate the entire inverse mapping in a black-box manner, we adopt a plug-andplay strategy in which a variational autoencoder (VAE) is used solely to learn prior information on conductivity distributions. During the inversion process, the conductivity model is iteratively updated using the Gauss Newton method, while the model space is constrained by projections onto the learned VAE decoder. This framework preserves explicit control over data misfit and enables flexible adaptation to different survey configurations. Numerical and field experiments demonstrate that the proposed approach effectively incorporates prior information, improves reconstruction accuracy, and exhibits good generalization performance. |
| title | Feature-based Inversion of 2.5D Controlled Source Electromagnetic Data using Generative Priors |
| topic | Geophysics Machine Learning |
| url | https://arxiv.org/abs/2601.02145 |