Diffracted arrivals are more sensitive to local heterogeneity than conventional reflections. However, the diffracted waves are usually buried in the reflection wave train, which makes them difficult to utilize. To address this challenge, we propose a workflow to separate the diffracted waves from reflections by using a deep learning network. The separated diffracted waves can be used to obtain a subsurface velocity model by using migration velocity analysis. The obtained velocity model allows us to migrate the original dataset into the final image. The workflow has been applied to a Ground Penetrating Radar (GPR) dataset to demonstrate the feasibility of the proposed workflow.

Presentation Date: Monday, October 12, 2020

Session Start Time: 1:50 PM

Presentation Time: 3:30 PM

Location: Poster Station 5

Presentation Type: Poster

Keywords:
deep learning, background velocity, geophysics, neural network, imaging, workflow, artificial intelligence, classification, imaging diffraction wave, diffraction wave
Subjects:
Information Management and Systems, Neural networks
This content is only available via PDF.
2020. Society of Exploration Geophysicists
You can access this article if you purchase or spend a download.