Iterative 3D Particle Reconstruction via Optical Transfer Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3D-PTV methods are limited by low particle density due to image overlapping, leading to inaccurate determination of particle positions and displacement vectors, while higher particle densities in Tomo-PIV methods are computationally intensive and time-consuming.
Innovation Solution
An iterative reconstruction method is introduced, where a rough initial particle distribution is iteratively refined by calculating virtual images based on optical transfer functions, comparing them to real images, and adjusting parameters until predetermined tolerance is met, eliminating the need for triangulation and reducing computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a larger number of detectors are used to increase particle density, then measurement precision improves, but device complexity and cost increase
Solution Approach 1:
The patent creates virtual images by copying and processing the real images through iterative reconstruction algorithms. Instead of adding more physical detectors, the system generates multiple virtual views of the particle distribution through computational methods, effectively simulating additional detection angles without requiring additional hardware detectors.
Solution Approach 2:
The patent changes the parameter of particle density in the measuring volume. By using iterative reconstruction with optical transfer functions, the system can accurately determine particle positions even at higher particle densities where traditional triangulation methods fail due to image overlapping. This allows operating in a previously inaccessible parameter regime.
2Measurement precision
If a larger number of detectors are used to increase particle density, then measurement precision improves, but computing time increases
Solution Approach 1:
The patent performs preliminary action by determining optical transfer functions for each detector before the actual iterative reconstruction process. These pre-determined transfer functions are then reused during the iterative refinement of particle distributions, avoiding the need to recalculate them at each iteration step and significantly reducing overall computing time.
Solution Approach 2:
The system creates virtual images through computational copying and processing of real images. By generating virtual views through iterative reconstruction using pre-determined optical transfer functions, the system achieves accurate particle position determination without requiring additional physical detectors or excessive computing resources.
3Productivity
If traditional triangulation is used with high particle density, then measurement speed is maintained, but measurement precision deteriorates due to image overlapping
Solution Approach 1:
The patent replaces the mechanical/geometric triangulation method with an iterative reconstruction approach based on optical transfer functions. Instead of relying on geometric intersections of particle positions from multiple detectors, the system uses iterative optimization to reconstruct the three-dimensional particle distribution, which is much more robust to image overlapping and higher particle densities.
Solution Approach 2:
The patent implements feedback through the iterative reconstruction process. The system repeatedly refines the estimated particle distribution by comparing virtual images generated from the current estimate with the actual real images, using the differences (feedback) to improve the estimate in subsequent iterations until convergence is achieved.
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 allows for higher particle densities and more accurate determination of displacement vectors without increasing computational time, improving the accuracy and efficiency of flow condition analysis.
Implementation Method 1
For each image detector: determination of an optical transfer function with which the real distribution is imaged by the image detector
Data Source
AI summary
The invention relates to a method for determining flow conditions in a measured volume permeated by a fluid spiked with optically detectable particles. A plurality of two-dimensional images of the particle distribution is thereby created at each of a plurality of times, an estimated particle distribution is determined therefrom, and a three-dimensional displacement vector field is calculated. According to the invention, a transfer function for the image detectors used is first determined, by means of which the real distribution is mapped by the image detector. Starting from a roughly estimated initial distribution, and by means of the transfer function, virtual images of the estimated distribution are then calculated and compared to the associated real images. The estimated distribution is modified in an iterative method until sufficient matching of the virtual and real images has been achieved.