Topological Knowledge Representation for Seismic Analogue Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Geoscientists face challenges in retrieving seismic images with specific topological structures, as traditional image search methods focus on visual attributes rather than structural arrangements, making it difficult to find analogues for petroleum exploration.
Innovation Solution
A method using topological knowledge representation (TKR) is employed, where a domain-specific knowledge base is utilized to build and validate a TKR input query, enabling statistical analysis to retrieve seismic images with similar geological structures by segmenting and classifying seismic datasets and matching them against a search database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image search methods are used to retrieve seismic images, then visual attribute matching is achieved, but topological structure matching capability is lost
Solution Approach 1:
The patent segments the seismic image into multiple regions and represents each region as a node in a topological graph. This segmentation allows the system to capture both visual attributes of individual regions and their topological relationships, resolving the contradiction by preserving topological structure information that would be lost in traditional holistic image search methods.
Solution Approach 2:
The patent transforms the image search problem from a two-dimensional visual attribute space into a topological graph space that incorporates spatial relationships and structural arrangements. This dimensional transformation enables simultaneous consideration of both visual attributes and topological structures, preventing loss of topological information while maintaining visual matching capability.
2Productivity
If traditional image search methods focus on visual attributes, then retrieval speed is maintained, but ability to find geological analogues with specific structures is reduced
Solution Approach 1:
The patent performs preliminary segmentation and topological graph construction on seismic images before the actual search process. By pre-processing images into structured topological representations, the system enables faster retrieval operations that can efficiently query structural characteristics without sacrificing geological analogue accuracy, thus maintaining both retrieval speed and reliability.
Solution Approach 2:
The patent changes the search parameters from purely visual attributes to include topological graph features such as node connections, regional arrangements, and structural relationships. This parameter transformation allows the retrieval system to efficiently search for geological analogues with specific structures while maintaining acceptable retrieval speed through optimized graph matching algorithms.
3Measurement precision
If seismic datasets are segmented and classified to build TKR queries, then structural graph extraction accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent introduces a topological graph as an intermediary representation between the segmented seismic regions and the final search query. This intermediary structure simplifies the processing complexity by providing a standardized framework for representing spatial relationships, making the system more manageable despite the increased accuracy requirements for segmentation and classification.
Data Source
AI summary
Method, apparatus, and computer program product are provided for retrieving analogues using topological knowledge representation (TKR). In some embodiments, a TKR input query is built and/or validated using a domain-specific knowledge base (KB). A search database containing candidate analogues and corresponding pre-built TKRs is then searched to retrieve at least one analogue of the TKR input query using statistical analysis. In some embodiments, a system may build the TKR input query based on a seismic dataset. For example, the system may receive a seismic dataset, segment the seismic dataset and classify each region using a computer vision (CV) database and the KB, and build the TKR input query based on the segmented and classified seismic dataset. In some embodiments, the TKR input query may be input and/or edited by a user. For example, the TKR input query may be input and/or edited by the user and validated using the KB.