Fracture Identification via Entropy Diffusion in Formation Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying and labeling fractures in subterranean formations is challenging due to their heterogeneity and discontinuous nature, requiring manual expertise and time-consuming processes, especially in 3D images, which hinders accurate petrophysical property analysis and hydrocarbon resource recovery.
Innovation Solution
An automated workflow using entropy filtering and diffusion equation solving to distinguish and label fractures in formation images, where entropy values are normalized and used as a diffusivity field to separate fracture elements from non-fracture elements, enabling efficient identification and labeling of fractures in both 2D and 3D images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual fracture identification by trained specialists is used, then measurement precision is improved, but productivity deteriorates
Solution Approach 1:
The system enables automated fracture identification where the image processing system performs the identification task itself without requiring continuous human intervention. The entropy filter and diffusion equation solver work autonomously to identify and label fractures, replacing the need for trained specialists to manually examine each image while maintaining identification accuracy.
Solution Approach 2:
The manual mechanical process of visual inspection by specialists is replaced with an automated computational system using entropy filtering and diffusion equations. The physical action of human eyes and brain processing images is substituted with mathematical algorithms that automatically detect fracture patterns based on entropy variations and diffusion characteristics.
2Productivity
If automated methods are used, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The system transforms the image data into entropy values and uses diffusion equations to enhance fracture visibility. By changing the parameter representation from raw pixel values to entropy-based diffusivity fields, the system automatically highlights fracture regions while suppressing background noise, enabling both speed and accuracy in automated identification.
Solution Approach 2:
The entropy filter acts as an intermediary transformation that converts standard image data into a diffusivity field where fractures are more distinguishable. This intermediate representation enhances the contrast between fracture and non-fracture regions, allowing automated algorithms to achieve high precision without manual intervention.
3Ease of operation
If traditional image processing is used, then ease of operation is maintained, but difficulty of detecting and measuring increases
Solution Approach 1:
The system automatically transforms images into entropy-based diffusivity fields, changing the parameter space to make fractures more detectable. This transformation simplifies the detection process by converting subtle fracture patterns into prominent high-entropy regions that are easier for automated systems to identify and measure accurately.
Data Source
AI summary
The disclosure provides an approach, or workflow, that extrapolates a segmentation carried out on a formation image into the labelling of fractures. The workflow can be applied to 2D and 3D images, which can be generated by different imaging technologies. Advantageously, one or more steps of the workflow can be performed automatically. An example of the workflow includes: (1) distinguishing fractures identified in a formation image from a background of the formation image by applying an entropy filter, wherein the formation image has elements that are defined as either fracture elements or non-fracture elements and entropy values for the elements are generated by the applying of the entropy filter, and (2) identifying the fracture elements that correspond to the fractures by solving a diffusion equation, wherein the entropy values are used as a diffusivity field for solving the diffusing equation and the fracture elements are used as a source.