Method of removing noise in seismic reverse-time migration

Patent metadata

Field	Value
Publication number	US-10281604-B2
Application number	US-201615068977-A
Country	US
Kind code	B2
Filing date	Mar 14, 2016
Priority date	May 20, 2015
Publication date	May 7, 2019
Grant date	May 7, 2019

Abstract

A method, including: obtaining, with a processor, a seismic image of a subsurface region from a computer memory; predicting, with a processor, a dip of the seismic image of the subsurface region; and removing, with a processor, noise or artifacts from the seismic image of the subsurface region by applying a dip guided Laplacian filter, wherein the removing generates another seismic image of the subsurface region that has noise or artifacts removed relative to the seismic image of the subsurface region.

First claim

The invention claimed is:

1. A method, comprising:
obtaining, with a processor, a stacked seismic image of a subsurface region from a computer memory;
predicting, with a processor, a dip of the stacked seismic image of the subsurface region; and
removing, with a processor, noise or artifacts from the stacked seismic image of the subsurface region by applying an image-dip-guided Laplacian filter along a normal direction of the stacked seismic image of the subsurface region, and
generating another stacked seismic image of the subsurface region that has noise or artifacts removed relative to the stacked seismic image of the subsurface region.

2. The method of claim 1 , wherein the obtaining further comprises:
dividing RTM input seismic data by frequency squared;
performing reverse-time migration by using source and receiving wave field cross-correlation imaging condition; and
generating the stacked seismic image of the subsurface region.

3. The method of claim 1 , wherein the removing includes using a mixed image-dip-guided Laplacian filter, which includes a weighted combination of a Laplacian filter and an image-dip-guided Laplacian filter.

4. The method of claim 3 , wherein the mixed image-dip-guided Laplacian filter includes a non-zero weighting parameter.

5. The method of claim 4 , wherein the weighting parameter has a value in a range of 0.1 to 0.5.

6. The method of claim 4 , wherein the weighting parameter has value greater than zero and less than or equal to one.

7. The method of claim 3 , wherein the weighting parameter is spatially dependent.

8. The method of claim 7 , wherein the weighting parameter is only depth dependent.

9. The method of claim 3 , wherein the removing includes applying the mixed image-dip-guided Laplacian filter multiple times, each with different values for the weighting parameter.

10. The method of claim 9 , wherein the predicting includes predicting image dip, partially or fully, by using a filtered image obtained in a previous iteration of the removing.

11. The method of claim 1 , wherein the predicting includes predicting the dip of the stacked seismic image of the subsurface region from the stacked seismic image of the subsurface region or from prior dip knowledge in the subsurface region, or a combination of both.

Assignees

• Jing Charlie
• Anderson John E
• Exxonmobil Upstream Res Co

Inventors

• Jing Charlie
• Anderson John E

Classifications

• G01V2210/324

  Filtering · CPC title

• G01V2210/522

  Dip move-out [DMO] · CPC title

• G01V2210/679

  Reverse-time modeling or coalescence modelling, i.e. starting from receivers · CPC title

• G01V2210/51

  Migration · CPC title

• G01V1/301

  for determining seismic cross-sections or geostructures · CPC title