... | ... | @@ -21,6 +21,10 @@ Hough transform |
|
|
|
|
|
Measurement of the particle flow
|
|
|
--------------------------------
|
|
|
When imaging accuracy is not high enough to detected particle Particle Image Velocimetry (PIV) can be performed. Check this with Agheal.
|
|
|
Low resolution imaging does not allow particle detection and finding their positions. However, displacement field of particles can be obtained using [Particle Image Velocimetry (PIV)](https://link.springer.com/article/10.1007/BF00190388). This method, by using correlations between neighboring cells in a coarse-grained grid, can calculate displacement fields between two images. The photo-elastic response images are not accurate enough for PIV. As the intensity of image changes drastically locally, it becomes very noisy for PIV to calculate displacements. To achieve a better precision, one use normal light images (without plorizers)to extract particle's trajectories. Several open-source packages in [Matlab](https://www.mathworks.com/matlabcentral/fileexchange/27659-pivlab-particle-image-velocimetry-piv-tool) or [other programming languages](http://www.openpiv.net/). Figure below shows a sample of PIV technique on unpplarized images. The blue arrows provide the displacement field of the granular medium under shear. The field represents data on a grid, showing mean flow around the points. The grid size chosen here is about the size of smaller particles for better accuracy. This technique can be used to extract local time series and mean flow of particles movements.
|
|
|
|
|
|
* The trajectories of particles obtained by PIV on unpolarized light images of a sheared granular medium. Blue arrows show the magnitude and direction of displacement field. The top image is an enlarged view of the bottom one:
|
|
|
![ch2_PIV](uploads/056de0a7beb1719b39e4238e382e6006/ch2_PIV.png)
|
|
|
|
|
|
|
|
|
[<go back to home](https://git-xen.lmgc.univ-montp2.fr/PhotoElasticity/Main/wikis/home) |
|
|
\ No newline at end of file |