Buffered Vision Frames

[oglakan]

This is more of a behind the scenes question rather than a coding one.  Let me give a setup before I ask.  I am using IMAQdx for a gige camera.  I am using a frame grabber in a while loop and processing the images, the details don't really matter.  The total time to do all the processing is about 200 ms, but I am running the camera at 30 fps.  Here comes the question.  Clearly my process is not running at 30 fps, but the camera is.  Will labview buffer the grabbed frames and then process them all, or will it do its process and grab the most current one once it goes to the next iteration.  I believe the later is the case, and that is what I want, I just wanted ask an expert to get what is going on under the covers.  Thx.

Matt

[Kristen_K. · Accepted solution]

Hi Matt!

You belief is correct.  With Grab Acquisition, one buffer is used and is overwritten.  You may want to try a Sequence Acquisiton, which uses the number of buffers defined by the user.  Take a look at the IMAQdx Help File, too. (Start >>Programs >> National Instruments >>Vision>>Documentation>>IMAQdx).   There's a page title Low-Level Function Programming Examples that has a pretty good overview of how Snap, Grab, and Sequence Acquisition work.  (I just searched "sequence").  I hope this helps!

Kristen H.

Message Edited by Kristen H. on 04-06-2009 03:52 PM