generic multi camera aquisition

[Jorge_Amaral]

Hello to all,

I'm a labview user for some years and I've a problem that I'm trying to solve for some time. This is related with the code reuse.

The purpose of my question is to find a way of building a framework that allows me to solve some simple vision tests with the few customizations. I hope that someone can point me a good direction. Because I can't.

My first approach (reentrant and call vi reference):

1. In my configurator I select the number of cameras that I want

2. I developed a vi that manages camera acquisition (error handling, grab, open, start, etc in a state machine). And made it reentrant.

2.1. The captured image is returned in a unique queue by camera (using the camera name to build the queue name).

3. The Main application only needs to read the camera names from the configuration file and using a for loop call (using "open vi reference")  N times the "cameraAquisition.vi".

This procedure works well when I the images do not need to be synchronized.

But in the project where this application was developed I needed to "glue" triggered images from 4 cameras to test the product. And this forced me to give up from the good idea that I explained. And I had to do the simple approach where I place 4 grab.vi in parallel.

So I started with the good intention of making a nice generic application (because our product can have many configurations possible) but finished with a solution for only one type of machine.

So... Can anyone give me some tips please. Forgive my simple labview concepts but I never had a Labview course. 

[falkpl]

This will work if your cameras support external triggering (most industrial cameras do).

Set all cameras to external striiering and generate a timed bulse from an external device (such as a counter).