Seismic Shear Noise Attenuation with Adaptive Curvelet Thresholding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for removing shear noise from seismic data, particularly in ocean bottom node environments, are inefficient and costly, often leading to loss of weak coherent signals and require manual intervention, with 3D Curvelet Transforms being computationally expensive.
Innovation Solution
An iterative approach using Fourier-Curvelet Transform with automatic thresholding, adaptive localized filtering, and optimum Bayesian weighting to attenuate shear noise, reducing computational burden and preserving signal integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If 3D Curvelet Transform is used for shear noise attenuation, then noise removal effectiveness is improved, but computational cost increases prohibitively
Solution Approach 1:
The patent segments the 3D Curvelet Transform into a two-step process: first applying 2D Curvelet Transform to reduce computational burden, then applying 1D Curvelet Transform along the remaining dimension. This segmentation allows the method to achieve effective 3D noise attenuation while avoiding the prohibitively high computational cost of direct 3D Curvelet Transform.
2Manufacturing precision
If manual thresholding values are defined through significant testing, then noise attenuation accuracy is improved, but processing time and complexity increase
Solution Approach 1:
The patent implements self-service by automatically defining thresholding values through statistical analysis of the Curvelet coefficients. The system uses the median absolute deviation (MAD) method to compute thresholds without requiring manual intervention or significant testing, thereby achieving accurate noise attenuation while dramatically reducing processing time and complexity.
3Manufacturing precision
If aggressive thresholding is applied to remove shear noise, then noise removal effectiveness is improved, but weak coherent signals are lost
Solution Approach 1:
The patent applies local quality by using adaptive thresholding that adjusts the thresholding strength based on the local characteristics of the data. The automatic threshold computation based on statistical properties (MAD) allows the method to apply stronger thresholding in noise-dominated regions while preserving weak coherent signals in signal-dominated regions, thereby achieving effective noise removal without losing important information.
4Use of energy by moving object
If 2D Curvelet Transform is used instead of 3D, then computational cost is reduced, but noise attenuation performance deteriorates
Solution Approach 1:
The patent segments the 3D noise attenuation problem into a sequential two-step process: first applying 2D Curvelet Transform to handle the majority of the computational workload and remove dominant noise components, then applying 1D Curvelet Transform along the remaining dimension to refine the attenuation. This segmentation achieves near-3D performance at 2D computational cost.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a method comprising: receiving captured seismic data and generating first velocity and first pressure data therefrom; transforming, from a first data domain to a second data domain, the first velocity and first pressure data and thereby generate second velocity and second pressure data; mapping wavenumber gather data from the second data domain to a third data domain; determining envelope ratio scaling data and threshold value data using the second velocity and second pressure data; generating, threshold envelope data using a thresholding operator associated with the threshold value data; using coefficient data associated with the threshold envelope data to estimate vertical component data; transforming, from the third data domain to the first data domain, the vertical component data to generate temporal-spatial data; and generating, based on the temporal-spatial data, noise component data comprised in the captured seismic data.