11-04-2013 04:02 AM
We have build a machine vision station and trying to sort pins by length as tall, short and medium. You can see the pins in the attached file. We tried line fitting algorithm, it calculates well but it is very slow. Can anyone suggest any better solutions to measure these pin lengths?
11-04-2013 05:02 AM
How about simple particle analysis? Threshold to binary, then particle filter followed by particle analysis. Max Feret Diameter particle measurement gives you the longest diameter which, depending if the differences between the lengths are very small or not, might be enough. If not, get particle Orientation and dynamically create a ROI for Gauge or alike.
Here's a snippet related to the former. Note that not all the stages here are necessary. In this case, second column of the array (Max Feret Diameter) would indicate the length.
11-04-2013 12:03 PM
Perhaps even simpler, simply use area instead of Max Feret Diameter.
11-05-2013 03:32 AM
Hi abilgi, You can also Equivalent Rect Long Side alone, since the pin is rectangular shape. for more on Particle measurement parameters, check the link.
http://zone.ni.com/reference/en-XX/help/372916L-01/nivisionconcepts/particle_measurements/
11-10-2013 01:48 PM - edited 11-10-2013 01:49 PM
you can find some examples of vision in