Skip to Main content Skip to Navigation

Interprétation géologique de données sismiques par une méthode supervisée basée sur la vision cognitive

Abstract : The goal of this work is building a new software platform for 3D seismic interpretation based on the geological knowledge currently used by domain experts. At present, seismic interpretation is generally based on image processing and seismic attributes. Results are interpreted by experts, who establish a correspondence between the image provided by the interpretation platform and a set of geological objects. We intend to explore here a new methodology based on cognitive vision, involving a formalization of the knowledge used for seismic interpretation. For this, we characterize the results of image processing operated over a seismic block, using concepts and visual relationships, which we describe in a specific ontology. This characterization enables us to formalize geological object interpretation so that interpretative solutions can be automatically proposed to experts for validation. Our work is focused on the identification of horizons and faults, which are the most basic objects entering in geological models. The software platform that we proposed is based on a specific ontology, that has been built for horizon and fault identification.
Document type :
Complete list of metadata
Contributor : Ecole Mines ParisTech Connect in order to contact the contributor
Submitted on : Wednesday, March 10, 2010 - 8:00:00 AM
Last modification on : Wednesday, November 17, 2021 - 12:31:12 PM
Long-term archiving on: : Friday, September 10, 2010 - 3:14:35 PM


  • HAL Id : pastel-00005861, version 1


Philippe Verney. Interprétation géologique de données sismiques par une méthode supervisée basée sur la vision cognitive. domain_other. École Nationale Supérieure des Mines de Paris, 2009. Français. ⟨NNT : 2009ENMP1666⟩. ⟨pastel-00005861⟩



Record views


Files downloads