Deep learning unflooding for robust subsalt waveform inversion

Full-waveform inversion (FWI), a popular technique that promises high-resolution models, has helped in improving the salt definition inverted velocity models. The success of relies heavily on having prior knowledge salt, and using advanced acquisition technology with long offsets low frequencies. Salt bodies are often constructed by recursively picking top bottom from seismic images corresponding to tomography combined flooding techniques. process is time-consuming highly prone error, especially (BoS). Many studies suggest performing FWI frequencies after constructing correct miss-interpreted boundaries. Here, we focus detecting BoS automatically utilizing deep learning tools. We specifically generate many random 1D containing or free bodies, calculate shot gathers. then apply starting flooded versions those results become inputs neural network, whereas true models output. network trained regression manner detect estimate subsalt velocity. analyze three scenarios creating training datasets test their performance 2D BP 2004 model. show when succeeds estimating velocity, requirement somewhat mitigated. In general, this work allows us merge top-to-bottom approach FWI, save time, empower converge absence data. This article protected copyright. All rights reserved