Improved impedance inversion by the iterated graph Laplacian

2024-04-25

AI Summary

To address low accuracy and poor robustness in seismic acoustic impedance inversion, this paper proposes a data-adaptive iterative graph Laplacian regularization method. Within a Tikhonov-type variational framework, the method integrates either classical (e.g., LSQR) or deep learning-based (CNN/UNet) initial estimates and dynamically constructs and recalibrates the graph Laplacian operator to embed structural priors throughout the inversion process—achieving unified design of initial-estimate guidance, graph-structure adaptation, and dynamic regularization parameter optimization. Experiments on synthetic and field seismic data under multiple noise levels demonstrate that, compared with single-initial-estimate methods, the proposed approach improves impedance estimation accuracy by 15–32%, accelerates convergence by 2–4×, and significantly enhances noise robustness.

Technology Category

• Reasoning under Uncertainty: Graphical Models
• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

We introduce a data-adaptive inversion method that integrates classical or deep learning-based approaches with iterative graph Laplacian regularization, specifically targeting acoustic impedance inversion - a critical task in seismic exploration. Our method initiates from an impedance estimate derived using either traditional inversion techniques or neural network-based methods. This initial estimate guides the construction of a graph Laplacian operator, effectively capturing structural characteristics of the impedance profile. Utilizing a Tikhonov-inspired variational framework with this graph-informed prior, our approach iteratively updates and refines the impedance estimate while continuously recalibrating the graph Laplacian. This iterative refinement shows rapid convergence, increased accuracy, and enhanced robustness to noise compared to initial reconstructions alone. Extensive validation performed on synthetic and real seismic datasets across varying noise levels confirms the effectiveness of our method. Performance evaluations include four initial inversion methods: two classical techniques and two neural networks - previously established in the literature.

Problem

• Improves acoustic impedance inversion in seismic exploration
• Integrates classical and deep learning methods with graph regularization
• Enhances accuracy and robustness against noise iteratively

Innovation

• Integrates classical or deep learning with graph Laplacian
• Iteratively refines impedance using graph-informed prior
• Enhances accuracy and robustness to noise