What is it about?
In exploration geophysics area, how to remove the ground-roll noise, which has a strong energy and conceals our useful seismic information, is an important data processing task. We brought up a novel approach that applies the Convolutional Neural Network to separate the ground-roll and reflections apart, and promising results were achieved to show the effectiveness of this method.
Featured Image
Photo by Zbynek Burival on Unsplash
Why is it important?
Ground-roll is large in energy than the reflections underground, thus makes our useful signal not easy to see. The method we proposed here can separate the ground-roll and make the reflections observable.
Perspectives
Hope this work can inspire more people to work on the application of deep learning, or to say, AI, in the area of geophysics exploration.
Zhuang Jia
Tsinghua University
Read the Original
This page is a summary of: Separating ground-roll from land seismic record via convolutional neural network, December 2018, Society of Exploration Geophysicists,
DOI: 10.1190/aiml2018-16.1.
You can read the full text:
Resources
Contributors
The following have contributed to this page
