Efficient 2-D Seismic Imaging via AI-Driven Frequency-Domain Interpolation and HPC Kernel Acceleration

Conference Paper (2026)
Author(s)

J. Zhao (The Cyprus Institute)

N. Akram (Universidade de Lisboa)

N. Savva (University of Cyprus)

E. Verschuur (TU Delft - Civil Engineering & Geosciences)

Research Group
Applied Geophysics and Petrophysics
URL related publication
https://www.earthdoc.org/content/papers/10.3997/2214-4609.202610613 Final published version
More Info
expand_more
Publication Year
2026
Language
English
Research Group
Applied Geophysics and Petrophysics
Volume number
2026
Article number
613
Publisher
European Association of Geoscientists & Engineers
Event
87th EAGE Annual Conference & Exhibition (2026-06-08 - 2026-06-11), Aberdeen, United Kingdom
Downloads counter
7
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

This research presents a hybrid AI–HPC acceleration framework designed to overcome the computational barriers of 2-D seismic imaging based on Full Wavefield Modeling. By exploiting the decoupling of frequency components, the proposed method reduces the computational burden of physics-based simulations through a sparse frequency sampling strategy. The omitted frequency slices are accurately reconstructed using an Attention U-Net, which employs attention mechanisms to filter noise and preserve structural details during the interpolation process. To further optimize runtime efficiency, the performance-hotspot wavefield propagation and scattering kernels are optimized via Numba-based Just-In-Time compilation. This implementation strategy eliminates the overhead of the Python interpreter, enabling native machine-code execution with SIMD vectorization and multi-core parallelism. Numerical experiments conducted on synthetic models with varying geological complexity confirm that this framework delivers substantial reductions in computation time while maintaining high wavefield accuracy, stable inversion convergence, and similar resolution in reflectivity imaging.

Files

613.pdf
(pdf | 1.5 Mb)
Taverne
warning

File under embargo until 11-12-2026