Particle Spatial Distribution Determination via Virtual Image Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the changing spatial distribution of particles at multiple points in time are inefficient due to high computing time and limited accuracy, particularly in Particle Tracking Velocimetry (PTV), as they rely on iterative triangulation and three-dimensional reconstructions from two-dimensional images.
Innovation Solution
A method that records real two-dimensional images with different mapping functions, calculates virtual images, determines differences, and varies the estimated spatial distribution based on shifts in particle locations between approximated distributions at previous points in time, reducing differences to achieve an accurate spatial distribution at each point in time with minimal computing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative triangulation and three-dimensional reconstructions are used to determine particle spatial distribution, then measurement precision is improved, but computing time increases significantly
Solution Approach 1:
The patent applies preliminary action by using particle positions from previous time points to pre-calculate expected positions at the current time point before actual measurement. This preliminary estimation provides a starting point that significantly reduces the number of iteration steps needed, thereby reducing computing time while maintaining measurement precision.
Solution Approach 2:
The patent uses virtual images as copies of real images to initialize the particle distribution. By creating virtual images from previous time point data and using them as starting estimates, the method avoids time-consuming triangulation while providing sufficient initial accuracy for the iterative optimization process.
2Measurement precision
If triangulation is used for three-dimensional reconstruction at each time point, then spatial distribution accuracy is improved, but the maximum particle density is limited due to application limits of triangulation
Solution Approach 1:
The patent changes the fundamental parameter from triangulation-based 3D reconstruction to a 2D image matching approach. By shifting from triangulation geometry to direct image comparison with virtual images, the method removes the particle density limitations inherent in triangulation while maintaining accurate particle location determination through the optimization process.
3Measurement precision
If conventional three-dimensional reconstruction methods are used, then complete spatial distribution is obtained, but device complexity and processing steps increase
Solution Approach 1:
The patent extracts and eliminates the complex triangulation and three-dimensional reconstruction steps from the measurement process. By directly working with 2D images from multiple cameras and comparing them with virtual images, the method achieves complete spatial distribution determination without the intermediate 3D reconstruction stage, thereby reducing processing complexity.
Data Source
AI summary
For determining a changing spatial distribution of particles at each of multiple points in time, real two-dimensional images of the particles are recorded with different mapping functions. An estimated spatial distribution of the particles is provided. Virtual two-dimensional images of the estimated spatial distribution are calculated applying the different mapping functions. Differences between the virtual and the real two-dimensional images are determined; and the estimated spatial distribution of the particles are varied for reducing the differences to obtain a spatial distribution approximated to the actual spatial distribution of the particles. The estimated spatial distribution of the particles is provided in that the locations of the individual particles in a spatial distribution approximated for one other point in time are shifted dependently on how the locations of the individual particles have changed between at least two spatial distributions approximated for at least two other points in time.