Hierarchical Multiresolution Segmentation for Seismic Data Volumes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for partitioning seismic data volumes into stratigraphically realistic layers or geobodies often result in unique decompositions that do not account for the multi-cyclical nature of stratigraphy, leading to inappropriate hierarchical representations.
Innovation Solution
A hierarchical multiresolution segmentation method that iteratively combines initial segments into larger units, allowing for the creation of nested layers or geobodies, using techniques such as size-guided, extrema-based, attribute-guided, and concurrence-based segment combinations to achieve stratigraphically realistic layering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If existing methods for partitioning seismic data volumes are used, then segmentation is achieved, but the decomposition does not account for the multi-cyclical nature of stratigraphy, leading to inappropriate hierarchical representations
Solution Approach 1:
The patent applies segmentation by dividing the seismic data volume into multiple hierarchical levels of segments. Each level represents a different scale of stratigraphic organization, allowing the multi-cyclical nature of stratigraphy to be captured through progressive partitioning from fine to coarse scales.
Solution Approach 2:
The patent implements nesting by creating hierarchical segments where smaller segments are contained within larger segments across different levels. This nested structure allows multiple stratigraphic cycles to be represented at different scales simultaneously, with each level nesting within the previous level to form a comprehensive hierarchical model.
2Manufacturing precision
If hierarchical multiresolution segmentation is applied, then geologically realistic geobodies are generated, but computational complexity increases
Solution Approach 1:
The patent applies dynamics by implementing an iterative segmentation process that progressively refines segments from coarse to fine scales. The method dynamically adjusts segment boundaries and characteristics at each hierarchical level, allowing computational resources to be focused on critical regions while maintaining geological realism.
Solution Approach 2:
The patent uses partial action by applying segmentation operations selectively at different hierarchical levels rather than uniformly processing the entire volume at maximum resolution. This allows the method to achieve geological realism in critical areas while reducing computational complexity in less critical regions.
3Measurement precision
If multiple segmentation techniques are combined, then segmentation accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by performing coarse-scale segmentation first to establish the overall hierarchical structure before applying finer-scale segmentation techniques. This preliminary partitioning reduces the computational burden of subsequent detailed segmentation operations while maintaining overall segmentation accuracy.
Solution Approach 2:
The patent uses periodic action by implementing segmentation at discrete hierarchical levels rather than continuously. The method progresses through periodic stages of refinement, applying different segmentation techniques at each level, which allows for efficient computation while maintaining accuracy through structured iterative processing.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
Method for segmenting a geophysical data volume such as a seismic data volume (10) for ranking, prioritization, visualization or other analysis of subsurface structure. The method takes any initial segmentation (11) of the data volume, and progressively reduces the number of segments by pair combination (13) so that an optimal stage of combination may be determined and used for analysis (15).