I reviewed 9 geophysics papers on Deep learning for Seismic INVERSE problems. |
|
In this video, I explain what is forward and inverse problems are, different conventional methods used for velocity model building (tomography, Full Waveform Inversion [FWI]), and go over 9 deep learning publications. #geophysics #oilandgas #research #geology
TIMESTAMPS 00:00 Introduction 01:16 Forward and Inverse problem 02:08 Estimating earth model 02:44 Tomography, FWI, MS-FWI 05:12 Into to Deep Learning 06:58 DL that improve FWI with Salt probability 07:44 DL that improve FWI with extrapolating low-frequency data 08:31 CNN for seismic impedance inversion 09:34 CNN for velocity model building 10:39 Encoder-Decoder for velocity model building 11:36 U-Net architecture for velocity model building 12:21 RNN for petrophysical property estimation from seismic data 13:25 Semi-supervised learning for acoustic impedance inversion 14:41 Wasserstein GAN for velocity model building 15:56 Pros and Cons of DL BE MY FRIEND: 🌎 My website/blog - https://ruslanmiftakhov.com/ 🪢 LinkedIn - https://www.linkedin.com/in/ruslan-miftakhov/ WHO AM I: I am Chief Technical Officer at a software company that develops AI/ML solutions for Oil and Gas sector. I am passionate about AI and its application in Oil and Gas. For 4-years, I have been learning, teaching, developing, and launching solutions based on AI for various aspects of geophysics, geomodeling, and petroleum engineering. I make videos about Deep Learning and Machine Learning applications within Oil and Gas. 📓 Resources Mentioned in the Video GitHub: • https://github.com/vishaldas/CNN_based_impedance_inversion • https://github.com/YangFangShu/FCNVMB-Deep-learning-based-seismic-velocity-model-building Papers: • Deep-learning tomography - https://www.researchgate.net/publication/322192763_Deep-learning_tomography • Applications of supervised deep learning for seismic interpretation and inversion - https://www.researchgate.net/publication/334314370_Applications_of_supervised_deep_learning_for_seismic_interpretation_and_inversion • Deep-Learning Inversion of Seismic Data - https://www.researchgate.net/publication/338036449_Deep-Learning_Inversion_of_Seismic_Data • Velocity model building with a modified fully convolutional network - https://www.researchgate.net/publication/327612425_Velocity_model_building_with_a_modified_fully_convolutional_network • Subsurface velocity inversion from deep learning-based data assimilation - https://www.sciencedirect.com/science/article/abs/pii/S0926985118308905 • Deep-learning inversion: a next generation seismic velocity-model building method - https://arxiv.org/abs/1902.06267 • Petrophysical Property Estimation from Seismic Data Using Recurrent Neural Networks - https://arxiv.org/abs/1901.08623 • Deep learning prior models from seismic images for full-waveform inversion - https://www.researchgate.net/publication/319162874_Deep_learning_prior_models_from_seismic_images_for_full-waveform_inversion • Extrapolated full waveform inversion with deep learning - https://arxiv.org/abs/1909.11536 • Deep learning for low-frequency extrapolation from multi-offset seismic data - https://www.researchgate.net/publication/335660119_Deep_learning_for_low-frequency_extrapolation_from_multi-offset_seismic_data • Seismic Full-Waveform Inversion Using Deep Learning Tools and Techniques - https://arxiv.org/abs/1801.07232 • PRE-STACK AND POST-STACK INVERSION USING A PHYSICS GUIDED CONVOLUTIONAL NEURAL NETWORK - https://www.jsg.utexas.edu/edger/files/Abstract_Biswas_2019.pdf • Semi-supervised Learning for Acoustic Impedance Inversion - https://arxiv.org/abs/1905.13412 • Stochastic seismic waveform inversion using generative adversarial networks as a geological prior - https://arxiv.org/abs/1806.03720 • Data-Driven Seismic Waveform Inversion: A Study on the Robustness and Generalization - https://ieeexplore.ieee.org/document/9044635 • Seismic impedance inversion based on cycle-consistent generative adversarial network - https://www.sciencedirect.com/science/article/pii/S1995822621000868 • Adler, A., Araya-Polo, M., & Poggio, T. (2021). Deep Learning for Seismic Inverse Problems: Toward the Acceleration of Geophysical Analysis Workflows. IEEE Signal Processing Magazine, 38(2), 89–119. doi:10.1109/msp.2020.3037429 BE MY FRIEND: 🌎 My website/blog - https://ruslanmiftakhov.com/ 🪢 LinkedIn - https://www.linkedin.com/in/ruslan-miftakhov/ WHO AM I: I am Chief Technical Officer at a software company that develops AI/ML solutions for the Oil and Gas sector. I am passionate about AI and its application in Oil and Gas. For many years, I have been developing and launching solutions based on AI for various aspects of geoscience and petroleum engineering. I make videos about Deep Learning and Machine Learning applications within Oil and Gas. |