English Russian


Use Machine Learning Techniques to Enrich the Data Available to Seismic Interpreters

August 06, 2019
This presentation is now available for viewing.

Presented by Sandra Allwork, Business Development Manager, and Camille Msika, Technical Sales Advisor
Domain:  Seismic Processing & Imaging
Products:  EarthStudy 360™, Rock Type Prediction



Machine Learning technologies are richly represented across the Emerson portfolio; here we focus on just two, which deliver enhanced data deliverables to aid the seismic interpreter. Deep Learning applied to full-azimuth data extracts ‘pre-classified’ data volumes from prestack data to highlight geological structures, while the Rock Type Classification workflow uses a Democratic Neural Network Association trained with both prestack data and well data to predict lithology distribution and probability. 


_IWM4715_sm.jpgSandra Allwork is Director of Technical Services at Emerson E&P Software in Europe, and has over 20 years of experience with the Paradigm portfolio. Over the years, Sandra has worked with customers in Europe and the Middle East to optimize their velocity model building and imaging workflows, incorporating the latest advances in anisotropy and full-azimuth analysis. She holds a BSc in Geology & Geophysics from the University of Durham in the UK.