Synthetic Data Generation for Deep Learning-Based Inversion for Velocity Model Building
Open Access
- 2 June 2023
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 15 (11) , 2901
Abstract
Recent years have seen deep learning (DL) architectures being leveraged for learning the nonlinear relationships across the parameters in seismic inversion problems in order to better analyse the subsurface, such as improved velocity model building (VMB). In this study, we focus on deep-learning-based inversion (DLI) for velocity model building, leveraging on a conditional generative adversarial network (PIX2PIX) with ResNet-9 as generator, as well as a comprehensive mathematical methodology for generating samples of multi-stratified heterogeneous velocity models for training the DLI architecture. We demonstrate that the proposed architecture can achieve state-of-the-art performance in reconstructing velocity models using only one seismic shot, thus reducing cost and computational complexity. We also demonstrate that the proposed solution is generalisable across linear multi-layer models, curved or folded structures, structures with salt bodies as well as higher-resolution structures built from geological images through quantitative and qualitative evaluation.Keywords
This publication has 38 references indexed in Scilit:
- A new optimization approach for source-encoding full-waveform inversionGeophysics, 2013
- The role of geophysics in petroleum resources estimation and classification—new industry guidance and best practicesThe Leading Edge, 2012
- Adjoint centroid-moment tensor inversionsGeophysical Journal International, 2011
- An overview of full-waveform inversion in exploration geophysicsGeophysics, 2009
- Three-dimensional frequency-domain full-waveform inversion with an iterative solverGeophysics, 2009
- Introduction: Towards a full waveform inversionGeophysical Prospecting, 2008
- Marmousi2: An elastic upgrade for MarmousiThe Leading Edge, 2006
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004
- LINEARIZED INVERSION OF SEISMIC REFLECTION DATA*Geophysical Prospecting, 1984
- Inversion of seismic reflection data in the acoustic approximationGeophysics, 1984