IP Library Granted Patent US 11,360,233
Granted Patent B2
US 11,360,233 · App. 16/646,910 · Granted Jun 14, 2022

Seismic image data interpretation system

Inventors: Mats Stivang Ramfjord (Oslo, NO); Nader Salman (Cambridge, MA); Michael Hermann Nickel (Tananger, NO); Guido van der Hoff (Tananger, NO)
Assignee: Schlumberger Technology Corporation
G01V1/50E21B7/04E21B47/0025E21B49/00E21B2200/20G01V99/00G01V2210/65G01V2210/66
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,360,233
App. No.
16/646,910
Granted
Jun 14, 2022
Kind
B2
Abstract

A method can include receiving seismic image data; processing the received seismic image data to generate stratigraphic information using a trained convolution neural network that includes channels subjected to convolution, activation and pooling that reduce spatial resolution and subjected to deconvolution and concatenation that increase spatial resolution; and enhancing the seismic image data using the stratigraphic information to generate an enhanced seismic image.

Claims (35)

1. A method comprising:

receiving seismic image data for a seismic image and depth information of a plurality of pixels in the seismic image, wherein the depth information is with respect to a reference location on the Earth;

processing, through a trained convolution neural network, the received seismic image data with the depth information to generate stratigraphic information, wherein the trained convolution neural network comprises channels subjected to convolution, activation and pooling that reduce spatial resolution and subjected to deconvolution and concatenation that increase spatial resolution, wherein the channels comprise a depth channel with the depth information; and

enhancing the seismic image data using the stratigraphic information to generate an enhanced seismic image.

2. The method of claim 1 , wherein the processing the received seismic image data comprises processing tiles of the seismic image data, wherein the tiles comprise a depth dimension that exceeds a lateral dimension.

3. The method of claim 1 , wherein the seismic image data comprises pixel image data.

4. The method of claim 1 , wherein the trained convolution neural network comprises a “U” shaped architecture.

5. The method of claim 1 , wherein the trained convolution neural network comprises at least two processes that reduce spatial resolution.

6. The method of claim 1 , wherein the trained convolution neural network comprises at least two processes that increase spatial resolution.

7. The method of claim 1 , further comprising:

training a convolution neural network to generate the trained convolution neural network, wherein the training comprises:

interpreting a portion of the seismic image data to generate labeled training data and processing the labeled training data to generate the trained convolution neural network, wherein generating the labeled training data comprises augmenting the portion of the seismic image data,

wherein the augmenting comprises generating synthetic labeled training data.

8. The method of claim 1 , comprising rendering the enhanced seismic image to a display.

9. The method of claim 1 , comprising locating a position in the enhanced seismic image that corresponds to hydrocarbons.

10. The method of claim 9 , comprising determining a drilling trajectory for a well that targets the position in the enhanced seismic image that corresponds to hydrocarbons.

11. The method of claim 1 , comprising using the enhanced seismic image in a drilling operation that drills a borehole to a position in the enhanced seismic image that corresponds to a reservoir.

12. The method of claim 1 , wherein the stratigraphic information comprises stratigraphic units.

13. The method of claim 1 , further comprising:

training a convolution neural network to generate the trained convolution neural network, wherein the training comprises interpreting a portion of the seismic image data to generate labeled training data and processing the labeled training data to generate the trained convolution neural network.

14. The method of claim 1 , further comprising:

training a convolution neural network to generate the trained convolution neural network, wherein the training comprises:

interpreting a portion of the seismic image data to generate labeled training data and processing the labeled training data to generate the trained convolution neural network, wherein generating the labeled training data comprises augmenting the portion of the seismic image data.

15. The method of claim 1 , wherein the depth information is on a per pixel basis of the seismic image.

16. A system comprising:

a processor;

memory operatively coupled to the processor; and

processor-executable instructions stored in the memory to instruct the system to:

receive seismic image data for a seismic image and depth information of a plurality of pixels in the seismic image, wherein the depth information is with respect to a reference location on the Earth;

process, through a trained convolution neural network, the received seismic image data with the depth information to generate stratigraphic information, wherein the trained convolution neural network comprises channels subjected to convolution, activation and pooling that reduce spatial resolution and subjected to deconvolution and concatenation that increase spatial resolution, wherein the channels comprise a depth channel with the depth information; and

enhance the seismic image data using the stratigraphic information to generate an enhanced seismic image.

17. One or more computer-readable storage media comprising computer-executable instructions executable to instruct a computing system to perform a method comprising:

receiving seismic image data for a seismic image and depth information of a plurality of pixels in the seismic image, wherein the depth information is with respect to a reference location on the Earth;

processing, through a trained convolution neural network, the received seismic image data with the depth information to generate stratigraphic information, wherein the trained convolution neural network comprises channels subjected to convolution, activation and pooling that reduce spatial resolution and subjected to deconvolution and concatenation that increase spatial resolution, wherein the channels comprise a depth channel with the depth information; and

enhancing the seismic image data using the stratigraphic information to generate an enhanced seismic image.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 16, 2020
From: RAMFJORD, MATS STIVANG; SALMAN, NADER; NICKEL, MICHAEL HERMANN; VAN DER HOFF, GUIDO
To: SCHLUMBERGER TECHNOLOGY CORPORATION
Reel/Frame 052118/0906 →
Continuity (2)
Provisional Application 62557746 · Sep 12, 2017
Related Publication 20200301036A1 · Sep 24, 2020
Cited By (3)
US 12,320,941 US 12,486,759 US 12,554,040