Best method for (simple) video acquisition and manipulation within Labview

[millonas]

We are doing a project where we would like to acquire and display a video image within Labview. In addition to this we want to be able to superimpose some graphics over the video image, if possible. For example, a circle or dot or some other marker that is superimposed over an individual moving through the video field. That is all at the moment - the guts of the application is (non-video) signal acquisition and analysis I will not need to analyze or otherwise use the raw video data, we just want to display the video and then place markers over the field.

I've seen some third party video packages highly recommended, but what I'd really like to know is what the best (and possibly cheapest) way to implement the two features described above.  I'm not a video person, and this will be my first time implementing this type of graphics within labview, so simple and high level would be best. I want to spend most of my time focusing on the real application, and the video is just an added feature I would like to work seamlessly.

Oh, I guess I should mention that we will be usng a standard USB video camera, nothing special or "homemade".

Thanks in advance for advice anyone can offer.  I hope this is in the right place. The app is not really "machine vision" but I was betting this forum would have the best knowledge base on this issue.

[Matt_McLaughlin]

Hi millonas,

This sort of task would be performed using LabVIEW's Vision Acquisition Software and the Vision Development Module. Do you already have this software available to you? If so, I would recommend working on developing the machine vision aspect of recognizing and circling an individual using Vision Assistant. You would then implement this processing to continuous image acquisition with inline processing in LabVIEW. I would recommend looking through the IMAQ examples in the Example Finder (Browse>>Hardware Input and Output>>IMAQ) for basic ideas on image processing in LabVIEW.

The USB camera will be supported by the IMAQdx drivers and their functions.

Please let us know if you would like further and more specific information on how these tasks would be performed. Thanks!

[millonas]

I think the vision acquisition module is ok. I will check it out.  However there is no machine vision component in this case as I mentioned. The "marker" is not based in any way on visual alnalys so the machine vision package is no help. The "position"  is inferred from a sensor net of an entirely differnet character.. Only certan people would become "targeted" in this way.  So the second question is probably simpler (and cheaper for us, lol)  than you made it out to be - is there a good way to place a graphaical object of our choice, such as a circle or dot, etc. at an arbirtary point (and move it around) on the fornt panel, especially in this case, IN FRONT of the vidoe display?

Thanks for the feedback.

[Matt_McLaughlin · Accepted solution]

Hi millonas,

As you suggested, the process of tracking an individual and determining where you want to place this graphical object will be much more complex than actually placing the object, but it sounds like you will be doing this elsewhere.

In regards to adding the shape to a front panel image display, you would likely want to use the VIs located in the IMAQ Overlay Sub-Palette. With these VIs you can add objects (arcs, rectangles, ovals, etc.) to your front panel image display. These functions are included with the Vision Acquisition Software, so you would not need the Vision Development Module to add these graphical objects to the image display.