MAU-Net:a multi-branch attention U-Net for full-wavefom inversion

Abstract

Data-driven velocity inversion has emerged as a prominent and challenging problem in seismic exploration. The complexity of the inversion problem and the limited data set make it difficult to ensure the stability and generalization of neural networks. To address these concerns, we propose a novel approach called multi-branch attention U-Net (MAU-Net) for velocity inversion. The key distinction of MAU-Net from previous data-driven approaches lies in its ability to not only learn information from the data domain, but also incorporate prior model domain information. MAU-Net consists of two branches: one branch uses seismic records as input to effectively learn the mapping relationship between the data and model domains, while the other branch employs a prior geological model as input to extract features from the model domain, thereby guiding MAU-Net’s learning process. Additionally, we introduce three major improvements in the model branching path to enhance MAU-Net’s utilization of seismic data and handle redundant information. We validate the effectiveness of each improvement through ablation experiments. The performance of MAU-Net is demonstrated with the Marmousi model and 2004 BP model, and it can also be combined with FWI to further improve the quality of the inversion result. MAU-Net exhibits robust performance on field data through the use of transfer learning techniques, further confirming its reliability and applicability.