From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.

We appreciate your patience as we improve our online experience.

LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Getting a signal (boolean) if the camera ROI detects movements

Introduction:

Hello, I'm pretty new to this camera acquisition thing and it really is easy to use and play with it, but I do not find a solution for this..

Project:

The project is for university, I need to do an automatic sorter so my idea was to take a photo from the conveyor, process it and use labview to control the motors and everything. I did almost everything from the acquisition image, but i can not find a way to get a boolean when something is entering in the ROI I made on a camera.
Problem:
How do I set a boolean true when the object is entering in the ROI?

Observations:

  • Because I need to take that photo to process it for optimization I will make the program to take a photo every time the object is entering in the ROI, not a photo every milisecond. 
  • I tried to use the color learn option, but I couldn't find a way to use this properly.
  • I can't use the tracking from vision assistant, because the object is not always in the frame.
0 Kudos
Message 1 of 4
(558 Views)

The "easiest" way to do this is to generate a TTL signal to trigger the "Snap" when the Object of Interest is in front of the camera.  Something like a photo-diode (what we used to call "an electric eye") watching the conveyer belt or a proximity sensor would do it.  You may have to play a bit with speed of the conveyor, size of the Object, size of the ROI, etc.

 

Bob Schor

Message 2 of 4
(528 Views)

That's a very good idea and I will take it into consideration, but I wonder.. Can't I make something in the software? 

I really thought it will be so easy and there must be an option/method to detect if something is changed in the ROI and now I'm too surprised that I can't find a solution.

Thanks you!

Adrian Geantă

0 Kudos
Message 3 of 4
(506 Views)

One of the most challenging LabVIEW "corner" is Image Processing -- not 1D, not 2D, but 3D data (two spatial dimensions, one intensity dimension which, if color, can be considered as RGB = 3 more dimensions).  I'm glad I had several years of LabVIEW experience before being "introduced" to LabVIEW Vision.

 

You could capture images with identified ROIs and figure out a "measure" to decide if is "sufficiently different".  Are you an IMAQdx expert?  Ready to spend a month or so learning?

 

Bob Schor

0 Kudos
Message 4 of 4
(489 Views)