A neural networks based seismic object detection technique
Authors:
- 1 January 2005
- conference paper
- Published by Society of Exploration Geophysicists
- Vol. 1 (1) , 775-778
Abstract
Seismic technology has historically focused on resolving the structures. To a lesser degree, presence or absence of reservoir has also been a focus. This paper shows the use of seismic data to detect several other “seismic objects”. The technique to be referred to as “meta attribute”, uses the combination of “artificial intelligence” of neural networks and the “natural intelligence” of an interpreter. Examples of many geologic features and reservoir properties detected using this technique will be provided. They include hydrocarbon probability, lithofacies, chimney, faults and salt.Keywords
This publication has 7 references indexed in Scilit:
- Geo-Hazard Detection with Chimney Cubes2003
- Meta-attributes—the key to multivolume, multiattribute interpretationThe Leading Edge, 2002
- Seismic inversion successfully predicts reservoir, porosity, and gas content in Ibhubesi Field, Orange Basin, South AfricaThe Leading Edge, 2002
- Identifying faults and gas chimneys using multiattributes and neural networksThe Leading Edge, 2001
- Quantitative Interpretation Generally Needs no Seismic Attributes1997
- Reservoir architecture and porosity distribution, Pegasus Field, West Texas — An integrated sequence stratigraphic‐seismic attribute study using neural networks1997
- Applications of clustering in exploration seismologyGeoexploration, 1984