Claims
- 1. A method of processing acquired seismic data which comprises extracting seismic information from the acquired data in a direction along the spatial direction of a body of interest thereby producing directional seismic attributes, characterized in that the directional attributes are combined using a supervised learning approach.
- 2. The method as claimed in claim 1, characterized in that the supervised learning approach comprises training an algorithm with a representative set of examples, by the steps of: identifying a set of points in a control volume of seismic data representative of a body and background; extracting selected directional attributes at these points and conveying these to the algorithm; and allowing the algorithm to learn how to combine the attributes, to provide an optimal classification into body and background.
- 3. The method as claimed in claim 2, characterized by applying the trained algorithm to a seismic volume, extracting directional attributes from every sample position in the seismic volume; and using the trained algorithm to classify the seismic volume in terms of body and background.
- 4. The method as claimed in claim 2, said algorithm comprises an artificial neural network.
Priority Claims (1)
Number |
Date |
Country |
Kind |
9819910 |
Sep 1998 |
GB |
|
RELATED APPLICATIONS
The present application claims the benefit of PCT Application No. PCT/GB99/03039, filed Sep. 13, 1999 and Great Britain No. Application 9819910.2, filed Sep. 11, 1998.
PCT Information
Filing Document |
Filing Date |
Country |
Kind |
PCT/GB99/03039 |
|
WO |
00 |
Publishing Document |
Publishing Date |
Country |
Kind |
WO00/16125 |
3/23/2000 |
WO |
A |
US Referenced Citations (7)
Foreign Referenced Citations (1)
Number |
Date |
Country |
9733184 |
Sep 1997 |
WO |
Non-Patent Literature Citations (2)
Entry |
Search Report—PCT/GB99/03039. |
Preliminary Examination Report—PCT/GB99/03039. |