07-27-2009 07:29 AM
07-27-2009
11:44 AM
- last edited on
04-24-2024
10:30 AM
by
Content Cleaner
Hi Hi_man,
If I understand you correctly, your inspection process will acquire images upon receiving one trigger and determine a pass/fail result. Then when a different trigger is hit, you will use that previous inspection result to control if the object should be removed. If this is correct, then this is fairly simple to do in LabVIEW on any of these platforms. In fact, if you are able to come to NIWeek in Austin next week we have a session entitled "Essential Techniques for Vision Timing and Synchronization" that covers this topic precisely.
In any case, you do not necessarily need multiple processors to accomplish this as LabVIEW is multithreaded and can do multiple loops at the same time. Multiple processors simply help speed things up. Since your other loop looking for your second trigger is not doing anything CPU intensive, it should be fine on any of these targets.
How you would write this depends on which platform you choose to use. If you use the NI 1764 smart camera, you would likely have your acquire/process loop put the results into a queue. Your second loop would use a WaitForSignal VI on an I/O and pull from the results queue when it gets the trigger, then generate a pulse if needed. On the CVS 1454 (or the new Embedded Vision System if you wanted something with more CPU power) you could use built-in FPGA to run the I/Os and queue the results.
Which platform you need will depend on the form factor, processing power, and type of camera you need to use for the inspection. If you need a self-contained camera and processing unit, then you'll want the smart camera. If you want ultimate flexibility in the camera you are choosing for the inspection, the CVS or EVS may be better. The NI 1764 Smart camera is a totally different architecture from the CVS or EVS (it has a PowerPC processor and a DSP), so the processing power cannot be directly compared to the CVS since it depends on the operation you are doing. The EVS on the other hand, with its larger form factor and higher power is obviously much faster than either of them.
Hope this helps,
Eric
07-31-2009 03:31 AM
Hi Eric,
As I see from your description, I can use smart camera to close my project and I would like to ask you one more question.
Can I use Labview to create a complicated algorithm and program the smart camera or I can only use Vision Builder to program smart camera.
Best regards,
Hi_man
08-01-2009 02:18 PM
Hi_man,
You can use either the LabVIEW Real-time Module or VBAI to target the Smart Camera.
-Christina