Machine Vision

cancel
Showing results for 
Search instead for 
Did you mean: 

Very basic pattern recognition

Hello,

Our company has a number of LabView and TestStand licenses, but we do not have the Ni Vision Development Module. Up until now it hasn't been necessary, and 

I don't think it is , alhough that is why I'm writing.

 

I need a very basic recognition solution. A device has Label A on the front and Label Be on the back. The device has a distinctive recognizable area on each side as well.

I need to look at the 'front' and distinguish Yes this the front and Yes this is Label A - No this isn't the front and No this isn't Label A

 

I would think there are inexpensive solutions for such a basic function, but I'm not having much luck finding them.

Any advice would be greatly appreciated.

0 Kudos
Message 1 of 3
(6,584 Views)
Hi Alan,
-I guess you need Vision Development module(VDM) for these pattern recognition.
-You can see lot of examples here using VDM http://www.ni.com/white-paper/6712/en/

-Looks like your application can be easily done by simple pattern matching and identifying the distinct recognizable area and decide which side it is,front or back.
-If you provide images, that would give better idea.
-But without VDM, i don't know how it can be done.
Thanks
uday
0 Kudos
Message 2 of 3
(6,582 Views)

Hello,

 

you could use OpenCV (http://opencv.org/) to achieve your goal, without needing the NI Vision Module. Build a dynamic link library (DLL) and call it in Labview. Simple pattern matching should be fairly simple to implement and would probably suffice in your case.

 

P.S.: You would need some C++ knowledge.

 

You can find some examples of calling OpenCV functions from Labview here:

 

https://decibel.ni.com/content/blogs/kl3m3n

 

It is used with the NI Vision Module, but only to display the image (I think no image processing is performed with NI Vision functions...)

 

Best regards,

K


https://decibel.ni.com/content/blogs/kl3m3n



"Kudos: Users may give one another Kudos on the forums for posts that they found particularly helpful or insightful."
0 Kudos
Message 3 of 3
(6,575 Views)