Experimental Setup

18 views

Skip to first unread message

Roja Najafi

unread,

Jul 29, 2025, 11:58:09 AMJul 29

to openpiv-users

I have a few questions regarding the experimental setup:

What kind of particles should be used?
I’m currently using small black particles like the ones in the link below. Their size is approximately 5 mm:
https://a.co/d/hLEYEhg
Contrast enhancement:
The contrast method from the OpenPIV tutorial works, but I’m experiencing some challenges related to background interference and light reflections.

In general, I’m facing two main difficulties:

Preprocessing the images
Opposite vectors appearing in the velocity field

I’ve attached 10 sample images so you can see the type of images I’m using.

Also, is there any recommended method for image preprocessing to retain the particles while masking or removing the background? I’ve tried basic thresholding and subtraction methods, but they haven’t worked well. So far, I’ve only applied thresholding followed by background subtraction.

I would appreciate any suggestions or guidance.

Best regards,

frame_00252.bmp

frame_00250.bmp

frame_00253.bmp

frame_00251.bmp

frame_00256.bmp

frame_00254.bmp

frame_00255.bmp

Alex Liberzon

unread,

Jul 29, 2025, 12:30:56 PMJul 29

to openpiv-users

Please explain first what you try to measure - are these particles floating? How deep is the channel? It appears that images are taken from above the channel, looking straight down; however, some particles are reflected from the glass below.

These images and these large black particles are not suitable for high-quality PIV analysis. It's better to either change the setup - using smaller particles and much denser seeding, or, if you have to use these particles, then it's better to perform particle tracking velocimetry. OpenPTV, MyPTV, proPTV, trackpy, etc., would do a better job here.