English Russian


Automatic Geological Feature Detection and Classification in the Imaged Directivity Domain

November 19, 2019
This presentation is now available for viewing.

Presented by:  Gali Dekel, Product Manager GeoDepth and EarthStudy 360
Domain: Processing & Imaging
Product:  EarthStudy 360™



The standard seismic image volume is dominated by the high energy specular data associated with principle reflectors and faults. Consequently, lower energy wavefields associated with stratigraphic features, reefs, and small faults are often lost in the standard processing and imaging process. 
This presentation shows an evolution of Emerson’s revolutionary full-azimuth imaging technology, performed in the Local Angle Domain, for characterizing subsurface features from migrated seismic data. The system decomposes the recorded seismic wavefield into full-azimuth directivity components comprised of thousands of dips and azimuths. We will demonstrate the use of Principle Component Analysis with its inherent data reduction, to derive the principle directivities.  Ultimately, we will use the power of deep learning to classify these directivities into geological features, such as reflectors and faults, plus other identifiable components, such as ambient noise or acquisition footprint. This is a reliable method for separating these components and producing targeted images from the composite wavefield.  The images reveal superior results over previous diffraction weighted stacks. Additionally, the deep learning approach offers significantly better time-to-results. 


Gali-Dekel_crop_sm-(1).jpgGali Dekel is the Emerson E&P Software Product Manager for GeoDepth and EarthStudy 360. She has worked for six years as a scientific developer and researcher in the company’s Processing & Imaging division. Gali holds a PhD in Physics from Tel-Aviv University.