EAGE London: AI-Driven Structural Seismic Interpretation
AI-Driven Structural Seismic Interpretation – application of ‘computer vision’ to complex interpretation challenges
The talk will have a hybrid form: in-person at Imperial College, London and broadcasted ONLINE.
Link to the webinar will be provided via e-mails to registered attendees: first e-mail will be sent two days before the event and the second one just 2 hours before the event. Do not register too late!
Lecture Room G38
Agenda (UK time)
18:30-18:35 Introduction and Announcements
18:35-19:20 AI-Driven Structural Seismic Interpretation – application of ‘computer vision’ to complex interpretation challenges
19:20-19:35 Q&A
from 19:35 Networking
Presenters
Dr Anat Canning, Distinguished Advisor | AI/ML, AspenTech
As Distinguished Technologist for Data Science at AspenTech, Dr. Canning leads the
research and development of machine learning technologies for next-generation E&P
solutions. Her expertise encompasses machine learning, imaging and inversion, amplitude
preservation, seismic-to-well ties, seismic anisotropy and the derivation of rock properties
from seismic. Prior to joining AspenTech, Dr. Canning worked at the Houston Advanced
Research Center (HARC) as a senior research scientist, and at the Institute for Petroleum
Research and Geophysics (IPRG) as a senior geophysicist. She has served as a lecturer at
leading technical universities worldwide.
Rob Bond, Advisory Consultant, AspenTech
Rob Bond is Advisory Solutions Consultant for Seismic Interpretation and Infrastructure in
AspenTech Europe. He has held various customer-facing technical, advisory and
managerial roles in Europe, Scandinavia, the Middle East and Asia Pacific, including a
twelve-year stint as Product Management Director for Seismic Interpretation. Rob holds a
First in Geological Sciences from Cambridge University. He is based in Surrey.
Talk outline
In recent years there has been a growing interest in the use of Machine Learning (ML)
technologies for processing and interpreting seismic data. Indeed, many procedures that
have traditionally been performed with deterministic methods and algorithms may now be
effectively replaced by Neural Networks and other AI methodologies, thereby improving
simplicity, productivity, and automation.
In this presentation, we will briefly review key aspects of this topic, discussing the use of
pre-trained neural networks versus training on individual datasets, strategies for data
labeling and “problem posing” used in building AI models, user prompts, and the relevant
implementations for solving specific interpretation tasks.
We will illustrate multi-horizon structural interpretation with examples of complex
unconformities and other cross-cutting geological features observed in seismic images,
together with results obtained from training a Deep Learning model based on an
interpreter’s interpretive perspective of selected seismic lines, and propagating those
supervised interpretations through large seismic volumes, with appropriate and rigorous
QC applied at all stages of the process.
AI-Driven Structural Seismic Interpretation – application of ‘computer vision’ to complex interpretation challenges
The talk will have a hybrid form: in-person at Imperial College, London and broadcasted ONLINE.
Link to the webinar will be provided via e-mails to registered attendees: first e-mail will be sent two days before the event and the second one just 2 hours before the event. Do not register too late!
Lecture Room G38
Agenda (UK time)
18:30-18:35 Introduction and Announcements
18:35-19:20 AI-Driven Structural Seismic Interpretation – application of ‘computer vision’ to complex interpretation challenges
19:20-19:35 Q&A
from 19:35 Networking
Presenters
Dr Anat Canning, Distinguished Advisor | AI/ML, AspenTech
As Distinguished Technologist for Data Science at AspenTech, Dr. Canning leads the
research and development of machine learning technologies for next-generation E&P
solutions. Her expertise encompasses machine learning, imaging and inversion, amplitude
preservation, seismic-to-well ties, seismic anisotropy and the derivation of rock properties
from seismic. Prior to joining AspenTech, Dr. Canning worked at the Houston Advanced
Research Center (HARC) as a senior research scientist, and at the Institute for Petroleum
Research and Geophysics (IPRG) as a senior geophysicist. She has served as a lecturer at
leading technical universities worldwide.
Rob Bond, Advisory Consultant, AspenTech
Rob Bond is Advisory Solutions Consultant for Seismic Interpretation and Infrastructure in
AspenTech Europe. He has held various customer-facing technical, advisory and
managerial roles in Europe, Scandinavia, the Middle East and Asia Pacific, including a
twelve-year stint as Product Management Director for Seismic Interpretation. Rob holds a
First in Geological Sciences from Cambridge University. He is based in Surrey.
Talk outline
In recent years there has been a growing interest in the use of Machine Learning (ML)
technologies for processing and interpreting seismic data. Indeed, many procedures that
have traditionally been performed with deterministic methods and algorithms may now be
effectively replaced by Neural Networks and other AI methodologies, thereby improving
simplicity, productivity, and automation.
In this presentation, we will briefly review key aspects of this topic, discussing the use of
pre-trained neural networks versus training on individual datasets, strategies for data
labeling and “problem posing” used in building AI models, user prompts, and the relevant
implementations for solving specific interpretation tasks.
We will illustrate multi-horizon structural interpretation with examples of complex
unconformities and other cross-cutting geological features observed in seismic images,
together with results obtained from training a Deep Learning model based on an
interpreter’s interpretive perspective of selected seismic lines, and propagating those
supervised interpretations through large seismic volumes, with appropriate and rigorous
QC applied at all stages of the process.
Good to know
Highlights
- 1 hour 30 minutes
- In-person
Location
Royal School of Mines
Royal School of Mines
London SW7 2AZ
How would you like to get there?
