Skip to main navigation Skip to search Skip to main content

Automatic salt detection with machine learning

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We introduce a novel method to estimate the shapes and positions of salt bodies directly from pre-stack seismic data using a modified fully convolutional network, which we use to perform both data transformation and semantic segmentation. Multiple shots are fed into the network as different channels to increase data redundancy. We generate synthetic data to train the network, and tests show satisfactory results.

Original languageEnglish
Title of host publication80th EAGE Conference and Exhibition 2018
Subtitle of host publicationOpportunities Presented by the Energy Transition
PublisherEuropean Association of Geoscientists and Engineers, EAGE
ISBN (Electronic)9789462822542
StatePublished - 2018
Event80th EAGE Conference and Exhibition 2018: Opportunities Presented by the Energy Transition - Copenhagen, Denmark
Duration: 11 Jun 201814 Jun 2018

Publication series

Name80th EAGE Conference and Exhibition 2018: Opportunities Presented by the Energy Transition

Conference

Conference80th EAGE Conference and Exhibition 2018: Opportunities Presented by the Energy Transition
Country/TerritoryDenmark
CityCopenhagen
Period11/06/1814/06/18

Fingerprint

Dive into the research topics of 'Automatic salt detection with machine learning'. Together they form a unique fingerprint.

Cite this