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.
Example Code
Products and Environment
This section reflects the products and operating system used to create the example.To download NI software, including the products shown below, visit ni.com/downloads.
- Vision Development Module
- LabVIEW
Software
Code and Documents
Attachment
Description
One fateful morning, I was pursuing the awesome LabVIEW example finder (Help >> Find Examples). I spotted an interesting application called “Optical Flow Feature Tracking Example.vi” which is installed with the LabVIEW Vision Development Module.
This example loads some high-speed video footage of an exploding firework, and tracks the movement of certain burning embers/particles between frames in the footage.
Wow! That looks seriously complex... or so I thought!
At the heart of the application was a pre-built function called IMAQ Optical Flow (LKP), which is included with the LabVIEW Vision Development Module. The rest of the code was architectural... *this* function represented the real “brains behind the operation”.
.JPG)
The IMAQ Optical Flow (LKP) function allows you to track the movement (flow) of certain features between frames in full motion video footage.
I was inspired! I decided to start developing my own example code that would work with live video footage, rather than a pre-recorded .avi. The attached project includes 3 different applications, each representing a different evolution of my own optical flow application.
1 Optical Motion (with velocity plot).vi
I gutted the example application, and added IMAQdx functions to pull live video feeds from my webcam. I also displayed the number of pixels a user-selected feature moves between frames (a direct representation of the speed that the feature is moving). I also added a new overlay that plots a vector to describe the velocity (speed & trajectory) of the moving feature.
To test drive the application, I drew a single dot on a piece of paper.
.jpg)
But the application had been written in such a way that it can simultaneously track any number of features.
.jpg)
I also decided to test drive the optical flow tracking on busier, more complex images. This worked brilliantly – illustrating how robust that IMAQ Optical Flow (LKP) function really is!!
.jpg)
2 Optical Motion (Overlay Eyes).vi
Following the success of the first iteration of the code, I decided to have some fun. This time I implemented features that allow you to add “Googly Eyes” to an object within the image frame. Far from being static, the Googly Eyes will actually stare at a user-controlled feature on the image (identified by a red crosshair).
.JPG)
Here is another image to prove that the object tracking is dynamic! No smoke and mirrors here 
.JPG)
Ofcourse, the scalable application architecture means that you can place eyes onto any number of people stood infront of the camera, and the algorithm will happily track the optical flow of multiple points – whilst ensuring that all eyes stare at the feature point!
.JPG)
And, it doesn’t just work on humans! The beauty of Googly Eyes is that they can be used to personify anything! Thereby enabling you to add a little charm and personality to just about any object...
.JPG)
3 Optical Motion (Looking For Geometry).vi
The final iteration of the application replaces the static feature point with a dynamic Geometry Match. I added the pre-built Geometry Matching Algorithms (that ship with the Vision Development Module) to the application. The code can be trained to search for any type of Geometry, by my example code looks for the following shape.
.JPG)
Each and every eye applied to you video feed will look directly at the detected geometry...
Your head moves = the pupils move!
The geometry moves = the pupils move!
.JPG){kind=tracking}
And, ofcourse, if you ever wanted to make yourself look like the Legendary Black Beast of Aaaaarrrrrrggghhh (from Monty Python and the Holy Grail), then this is the app for you!
.JPG)
To run this application on your own machine; you will need a copy of LabVIEW 2012, the Vision Development Module 2012 and the Vision Acquisition Software 2012.08 (or above)
You will also need a camera. You can use any direct-show compliant USB webcam, Firewire (IEEE 1394) camera or GIGE camera.
Important: I wrote the code for a colour camera...
if you are using a monochrome/greyscale camera, please remove
the “IMAQ Extract Single Color Plane” function from the block diagram!
Embrace the Silliness!
Example code from the Example Code Exchange in the NI Community is licensed with the MIT license.
Rich that's such a cool example! How did you create the effect at the very end of the video where you wipe all of the eyes away?
Also, are the Googly Eyes ripped out of Teasing Homer Simpson? 
Great example 
I have tried every example. But I am in trouble with the last one. I do not know how to set the option parameter for the IMAQ Optical Flow VI for my set camera. I have an error as an ouput of this VI because of the size.
Any ideas?
Thanks for your help!
