Published September 4, 2023 | Version v1
Conference paper Open

Deep Learning for the Inversion of Airborne EM Data

  • 1. The University of British Columbia, Dept of EOAS, UBC, Vancouver, BC., ehaber@eoas.ubc.ca
  • 2. Computational Geoscience Inc, christoph@compgeoinc.com

Description

In the recent decade Deep Learning have revolutionised fields such as computer vision and image understanding. However, its use for the solution of inverse problems have been limited. In this work we examine the use of deep learning for the processing and inversion of airborne EM data. Preliminary results show that by incorporating deep learning it is possible to eliminate many of the artefacts that are commonly observed in airborne inversion allowing us to obtain much more reliable inversions that fit not only the data, but also our a-priori information.

Other

Open-Access Online Publication: October 30, 2023

Files

AEM2023_ID016.pdf

Files (202.5 kB)

Name Size Download all
md5:0008957f72e65b0e18d8139703c690a2
202.5 kB Preview Download