2014 · 12 citations · 13 references
Natural SciencesEarth SciencesGeophysicsMarine GeologyMarine GeophysicsSeismologyGeological ModelingGeophysical Data ProcessingGeophysical InversionGeophysical Signal ProcessingSeismic Data ProcessingSeismic ImagingAnisotropic DiffusionComputational Inverse ProblemsFull-waveform InversionIll-posed Problem RegularizationRegularizationRegularization TechniquesRegularized InversionSeismic Inversion
Summary Full-waveform inversion (FWI) has now become a standard part of the workflow for building earth-models in the industry given its ability to provide high-resolution high-fidelity earth-models especially in complex geological settings. Inverse problems like FWI are ill-posed and ill-conditioned, which means that their solution is not unique and unstable. Some kind of regularization, usually in the form of smoothing of the model updates, is essential to obtain a stable solution. The most common approaches used in FWI previously, have been ad hoc in the sense that the smoothing parameters have not been influenced by the geology of the subsurface, which results in smoothing across major geological interfaces, such as, between salt and sediment. Here we present a method to regularize the FWI problem using an anisotropic diffusion-based regularization scheme to conform the earth model updates along geological structure automatically derived from a seismic image. We apply this method on a 3D offshore Brazil field dataset and show that it leads to a more geologically plausible model.
13
Solutions of Ill-Posed Problems.
John B. Bell, V. Y. Arsenin · Mathematics of Computation · 1978 · 8.2K citations
An overview of full-waveform inversion in exploration geophysics
J. Virieux, S. Operto · Geophysics · 2009 · 3.6K citations
Nonlinear anisotropic filtering of MRI data
Guido Gerig, O. Kübler, Ron Kikinis et al. · IEEE Transactions on Medical Imaging · 1992 · 1.2K citations
Coherence-Enhancing Diffusion Filtering
Joachim Weickert · International Journal of Computer Vision · 1999 · 767 citations