3D Integration Operator for Seismic Image Orientation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional seismic processing techniques lack adequate horizontal and vertical resolution to identify subtle geologic features such as small faults, fractures, and channels, which are crucial for hydrocarbon reservoir characterization, and are sensitive to noise and operator length, failing to capture dipping, azimuth, and volume change information in 3D seismic images.
Innovation Solution
A new 3D integration operator is introduced, precomputed for each x, y, and z dimension with a given operator length, applied to 3D post-stack seismic data to generate filtered data that enhances resolution, reduces noise, and captures essential orientation information, extending 2D integration methods into three dimensions using Lagrange interpolation and double integration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional seismic processing techniques are used, then processing simplicity is maintained, but horizontal and vertical resolution is inadequate to identify subtle geologic features
Solution Approach 1:
The patent extends traditional 2D integration operators into three dimensions by developing a 3D integration operator that operates on volumetric seismic data. This dimensional extension enables simultaneous filtering along x, y, and z axes, capturing dipping, azimuth, and volume change information that 2D operators miss, thereby improving resolution without proportionally increasing complexity
Solution Approach 2:
The 3D integration operator is implemented by precomputing separate filter masks for each of the three dimensions (x, y, and z), then applying them independently to the seismic data volume. This segmentation allows the complex 3D operation to be broken into manageable components that can be processed efficiently through systematic application to 3D-sub-cubes
2Measurement precision
If gradient operators are used for orientation analysis, then edge detection capability is provided, but sensitivity to noise and lack of flexibility with operator length occurs
Solution Approach 1:
The patent introduces integration operators as an intermediary approach between direct gradient computation and final orientation estimation. By first applying integration to smooth the data and reduce noise impact, then performing gradient analysis on the integrated results, the method achieves more reliable orientation estimates that are less sensitive to noise while maintaining accuracy
Solution Approach 2:
The integration operator provides flexibility in operator length selection, allowing adjustment of the filtering window size to match the scale of geological features of interest. This parameter flexibility enables optimization of the balance between noise reduction and feature preservation, improving reliability across different seismic data conditions
3Measurement precision
If 2D integration is applied, then local orientation estimation accuracy improves with reduced noise amplification, but dipping and azimuth information in 3D is not captured
Solution Approach 1:
The patent transitions from 2D integration operators to 3D integration operators that operate on volumetric seismic data. This dimensional extension enables the operator to capture dipping, azimuth, and volume change information by integrating along all three spatial dimensions, preventing information loss while maintaining the noise reduction and accuracy benefits of integration
Data Source
Figure 1~2
Figure 3
Figure 4A
AI summary
A separate three-dimensional (3D) integration filter mask if precomputed for each of x, y, and z dimensions with a given operator length. A portion of a 3D post- stack seismic data set is received for processing and loaded into a generated 3D-sub- cube. The separate 3D integration filter masks are applied to the loaded 3D-sub-cube to generate filtered 3D-sub-cube data. The square mean of the 3D-sub-cube is calculated to generate smoothed 3D-sub-cube data.