| Speaker: | Johnathan Kuttai, Ph.D Candidate University of British Columbia |
| Title: | An automated approach to incorporate structural information into the inversion using image segmentation |
| Date: | Tues, December 17, 2024 |
| Time: | 4:00pm to 4:40pm PST |
| Location: | Room 111 – 409 Granville Street Vancouver, BC, V6C 1T2 |
Abstract
Geological structure is distinctive and boundaries are sharp contacts between units. On the other hand, geophysical models are often smooth and geological meaning interpreted. These interpretations are often subjective and the model can be unconstrained and not always influenced by prior information. Presenting these results to a non-geophysicist can be challenging. Typically, delineations are sketched or overlain on the geophysical model to communicate the results thoroughly. Information on structure orientation is often ambiguous. With prior information, we can select more geologically relevant models using regularization. However, it is more common that prior information is not available. This work proposes an automated approach to infer structure in geophysical models. This is done by applying image segmentation methods borrowed from computer vision. This work uses a Transformer network (Vaswani, 2017) to perform the segmentation. The segmentation then becomes the prior information that we update the regularization with as we iterate an inversion. Within the regularization, we can indicate spaces in the model and the directions to regularize. We also explore additional methods to incorporate segmentation to the inversion via proximal operators and the Alternating Method of Multipliers (ADMM). A more clearly defined targeting system for interfaces like unconformities or structural orientations of intrusive units, which are often associated with mineralization zones, is provided by incorporating segmentation methods. The results provide an outlook on using machine learning and foundational models for structural interpretation of geophysical models.
Bio
Johnathan Kuttai graduated from the University of Saskatchewan with a B.Sc in geophysics in 2010 and began his career in data acquisition. Johnathan then joined an instrumentation group, writing software for signal processing large-scale DC resistivity and induced polarization data. During his career, he gained experience processing and acquiring many km’s of electromagnetic and potential field data worldwide in various environments, cold to hot, flat to mountainous, and desert to jungle. In 2017, Johnathan joined the SimPEG group to explore the geophysical inversion space, which eventually led to an interest in pursuing a Ph.D. Johnathan is currently with the UBC-GIF group, researching avenues in geophysical inversion and machine learning. Johnathan also works part-time doing inversion and research & development at Computational Geosciences Inc.
Recording