Acoustic-Sample Image Alignment for Drilling Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
On-site analysis of structural features in drilled or excavated samples is often tedious and time-consuming, prone to operator errors, and struggles to accurately correlate depth levels with physical features in images.
Innovation Solution
A machine learning model aligns acoustic images with sample images using orientation data to generate a virtual orientation line, enabling the determination of sample orientation and identifying physical features like fractures and bedding within the samples.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated image analysis techniques are used to analyze sample images, then analysis time is reduced and productivity is improved, but measurement precision deteriorates due to operator error when samples are improperly handled and inability to correlate depth levels to physical features
Solution Approach 1:
The patent introduces an intermediary alignment process that matches acoustic image features with sample image features to establish depth level correlations. This intermediary step bridges the gap between automated image analysis and accurate depth measurement, allowing the system to maintain high productivity while restoring measurement precision through feature-based registration rather than direct manual measurement.
Solution Approach 2:
The patent replaces manual mechanical measurement methods with an automated image processing system that uses acoustic and optical image alignment. By substituting the mechanical measurement process with computational image analysis, the system achieves both improved productivity and maintained measurement precision through digital correlation of depth levels with physical features.
2Measurement precision
If manual analysis methods are used to determine structural features, then measurement precision is maintained through direct observation, but productivity deteriorates due to tedious and time-consuming labor
Solution Approach 1:
The patent implements a self-service automated system that performs structural feature identification without requiring manual intervention. The machine learning model automatically detects and characterizes structural features in sample images, eliminating the need for tedious manual analysis while maintaining measurement precision through algorithmic pattern recognition that replicates expert observational capabilities.
3Device complexity
If traditional image analysis without acoustic correlation is used, then device complexity is reduced, but loss of information occurs regarding depth levels and spatial orientation of physical features
Solution Approach 1:
The patent merges acoustic imaging data with optical sample image data into a unified correlated visualization. By combining these two information sources through feature matching and alignment, the system recovers depth level information and spatial orientation data that would otherwise be lost, while maintaining relatively simple device architecture through integrated processing rather than separate complex systems.
Data Source
AI summary
Provided herein are methods and systems for improved acoustic data and sample analysis. A machine learning model may align an image of a sample with an acoustic image associated with the sample. The alignment of the image of the sample with the acoustic image may be used to generate a virtual orientation line. An output image comprising the virtual orientation line and the image of the sample may be generated. The output image may be displayed at a user interface that allows a user to interact with the output image.