Multi-Scale Rock Fabric Characterization via Nested Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining rock properties are limited by the small field of view provided by digital rock physics imaging techniques, making it difficult to accurately characterize large-scale rock samples efficiently and economically, which is crucial for hydrocarbon exploration and production.
Innovation Solution
A method that involves detecting texture descriptors, extracting features, and using pattern recognition to classify and segment rock fabrics, enabling the upscaling of petrophysical properties from small-scale to large-scale samples through multi-scale imaging and spatial optimization of key points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If digital rock physics imaging techniques are used to analyze rock properties, then measurement precision is improved, but the field of view is limited to small scales
Solution Approach 1:
The patent implements a multi-scale imaging approach where images at different magnification levels are nested within each other. A low-magnitude overview image provides the broad field of view, while high-magnitude detailed images are embedded at specific locations within the overview. This nesting allows the system to simultaneously access both large-scale spatial context and fine-scale rock property details, resolving the contradiction between field of view and measurement precision.
Solution Approach 2:
The patent segments the rock sample analysis into multiple scale levels. Instead of attempting to capture the entire sample at one uniform scale, the system divides the analysis into overview regions and detailed regions of interest. This segmentation allows different portions of the sample to be imaged at appropriate scales, enabling both broad coverage and high-precision measurement where needed.
2Reliability
If large-scale rock samples are imaged to capture comprehensive rock properties, then reliability of reservoir characterization is improved, but resource and time expenses increase
Solution Approach 1:
The patent performs preliminary low-magnitude imaging to generate an overview of the entire rock sample before conducting detailed high-magnitude imaging. This preliminary action identifies regions of interest and allows the system to plan subsequent detailed imaging strategically. By preparing the overview first, the system avoids time-consuming full-sample high-resolution imaging while still ensuring comprehensive reservoir characterization through targeted detailed analysis.
Solution Approach 2:
The patent applies partial action by imaging only the necessary portions of the rock sample at high resolution. Instead of uniformly imaging the entire sample at maximum detail, the system performs detailed imaging only in regions identified as important in the overview, such as areas containing critical pore structures or mineral variations. This partial imaging approach maintains characterization reliability while significantly reducing time and resource expenses.
3Area of stationary object
If multiple high-resolution images are stitched to create large field of view, then area coverage is improved, but device complexity and processing requirements become infeasible
Solution Approach 1:
The patent resolves the complexity issue by changing the organizational dimension from a flat 2D mosaic to a hierarchical 3D structure. Instead of stitching images side-by-side in a single plane, the system organizes images in a multi-level hierarchy where overview images contain references to detailed images, which in turn may contain even finer details. This dimensional transformation reduces processing complexity by allowing the system to work with compressed hierarchical representations rather than managing vast numbers of individual image stitches.
Data Source
AI summary
A method for determining fabric and upscaled properties of a geological sample, such as a rock sample. A system for the method also is provided.


