What is it about?
High-resolution seismic migration images are important for geophysical interpreters for characterizing the underground reserviors. However, the traditional least-squares migration method requires expensive computational and storage costs. We proposed a point spread function deconvolution method based on deep learning. Compared with the traditional least-squares migration method and the conventional deblurring filter method, our technique achives a better deconvolution result with reduced computational costs.
Featured Image
Photo by NASA on Unsplash
Why is it important?
Our findings show that we could enhance the resolution of migration images with sufficiently reduced computational costs compared with the traditional least-squares reverse time migration method.
Perspectives
Writing this article was excited as our method can achieve good results in terms of results, and it is also promising to apply our method to industrial production.
Cewen Liu
Tsinghua University
Read the Original
This page is a summary of: Deep learning-based point-spread function deconvolution for migration image deblurring, Geophysics, June 2022, Society of Exploration Geophysicists, DOI: 10.1190/geo2020-0904.1.
You can read the full text:
Resources
Contributors
The following have contributed to this page
