Deep Compressed Learning for 3D Seismic Inversion
This addresses the problem of costly data acquisition in seismic imaging for geophysics, though it appears incremental as it builds on existing compressed-learning frameworks.
The paper tackles 3D seismic inversion from pre-stack data with very few seismic sources by combining compressed sensing and machine learning in an end-to-end deep compressed learning approach, achieving an order-of-magnitude reduction in seismic records used during training while maintaining comparable 3D reconstruction quality.
We consider the problem of 3D seismic inversion from pre-stack data using a very small number of seismic sources. The proposed solution is based on a combination of compressed-sensing and machine learning frameworks, known as compressed-learning. The solution jointly optimizes a dimensionality reduction operator and a 3D inversion encoder-decoder implemented by a deep convolutional neural network (DCNN). Dimensionality reduction is achieved by learning a sparse binary sensing layer that selects a small subset of the available sources, then the selected data is fed to a DCNN to complete the regression task. The end-to-end learning process provides a reduction by an order-of-magnitude in the number of seismic records used during training, while preserving the 3D reconstruction quality comparable to that obtained by using the entire dataset.