3D Particle Segmentation Using Dual Seeding for Mixed Grain Sizes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging modalities face challenges in accurately segmenting samples with small grains or powders containing a range of particle sizes, particularly in non-destructive imaging systems like x-ray CT microscopy.
Innovation Solution
A method involving binarization, distance transform, marker creation, and dilation is employed to segment samples into unique particle classifications, with separate processing for different particle sizes, and a composite image is created by overlaying images for smaller and larger particles, followed by thresholding and H-max algorithm application for small particles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a single segmentation approach is used for all particle sizes, then the processing is simple, but the segmentation accuracy deteriorates for samples with mixed particle sizes
Solution Approach 1:
The patent applies segmentation by dividing the particle analysis into multiple size-based groups. Different segmentation parameters and algorithms are applied to each group: larger particles use one set of parameters while smaller particles use another. This resolves the contradiction by maintaining simplicity within each size group while achieving high accuracy across the full range of particle sizes through the segmented approach.
2Measurement precision
If different processing methods are applied to different particle sizes, then the segmentation accuracy improves, but the processing complexity increases
Solution Approach 1:
The patent implements local quality by applying different segmentation parameters and algorithms to different particle size ranges. Each size group receives optimized processing tailored to its specific characteristics, improving accuracy for each group. The overall complexity is managed by organizing these different processing methods into a structured workflow with clear size thresholds and classification criteria.
3Productivity
If small particles are analyzed with the same method as large particles, then the processing is efficient, but the measurement precision deteriorates for small particles
Solution Approach 1:
The patent applies dynamics by adaptively selecting different processing methods based on particle size. The system dynamically adjusts segmentation parameters, algorithm selection, and processing intensity according to the detected particle size range. This allows efficient processing of large particles while applying more sophisticated methods to small particles when needed, optimizing both productivity and precision for each size category.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively segments samples into distinct particles, allowing for accurate measurement of characteristics such as minor and major axis length, total volume, and sphericity, enhancing the analysis of grain characteristics and mineral content in 3D mineral maps.
Implementation Method 1
The X-ray CT microscopy systems irradiate the sample with x-rays, typically in a range between 1 and several hundred keV
Implementation Method 2
x-ray attenuation coefficients are calculated for the defined minerals
Data Source
AI summary
A multi scale material segmentation method is provided that creates markers to identify unique particles, for small and large particles independently, and then separately processes those markers.