pixel locations

[SMRLAB]

I am new to LabView and NI Vision. Currently I am using NI Vision 8.2 and LabView 8.5. I am looking for ideas to get the pixel locations of where the circles intersect the 8 lines. I need these locations to compare their distance from an origin. Attached is the picture I am working with. I am not sure if pattern matching or another VI would serve my purpose best.

Thanks,

[BruceAmmons · Accepted solution]

I would think pattern matching would work very well to find the intersections.  The only problem would be that you would have no idea which one is which, so you would need to come up with a sorting algorithm.  If you can locate the center (fairly easy if you find the big black area in the center using thresholding), you can calculate the distance to center and the angle from center.  Group by angle, then sort by distance to get them all in a known order.  Make sure you account for the possibility of missing points.

What are you doing with the target?  Measuring distortion?  Measuring magnification?  I did some work with side view mirrors a long time ago (my first vision project, believe it or not), so I am curious.  We were measuring vibration characteristics as we shook the mirror.

Bruce

[SMRLAB]

Bruce,

Thanks for the advice! I just wanted to be sure that I was heading down the right path. This is a distortion test that I am trying to upgrade. The glass is static and the radius of each circle is measured at each intersecting point. Its seems like a fairly simple program and I figured it would be a good learning experience for NI Vision.

Mark