Application of deep learning for seismic horizon interpretation
James Lowell A C and Jacob Smith BA GeoTeric, Generator Studios, Trafalgar Street, Newcastle Upon Tyne, UK, NE1 2LA.
B Level 14 Governor, Stirling Tower, 197 St Georges Terrace, Perth, WA 6000, Australia.
C Corresponding author. Email: james.lowell@geoteric.com
The APPEA Journal 59(1) 426-431 https://doi.org/10.1071/AJ18163
Submitted: 10 January 2019 Accepted: 26 February 2019 Published: 17 June 2019
Abstract
The interpretation of key horizons on seismic data is an essential but time-consuming part of the subsurface workflow. This is compounded when surfaces need to be re-interpreted on variations of the same data, such as angle stacks, 4D data, or reprocessed data. Deep learning networks, which are a subset of machine learning, have the potential to automate this reinterpretation process, and significantly increase the efficiency of the subsurface workflow. This study investigates whether a deep learning network can learn from a single horizon interpretation in order to identify that event in a different version of the same data. The results were largely successful with the target horizon correctly identified in an alternative offset stack, and was correctly repositioned in areas where there was misalignment between the training data and the test data.
Keywords: artificial intelligence, machine learning.
James Lowell holds a PhD in machine learning and medical image processing from the University of Durham. He joined GeoTeric in September 2007, specialising in R&D and Intellectual Property as well as supporting the GeoTeric software development team. With over 15 years’ experience in research and development he became R&D Director in January 2018. |
Jacob Smith joined GeoTeric in January 2015 as a Geoscientist in Perth, Australia and became Senior Geoscientist in July 2016. He holds a BSc (Honours) in Geophysics from the Curtin University of Technology and has over 12 years of industry experience, including roles with Schlumberger and CGG. |
