Machine Vision

cancel
Showing results for 
Search instead for 
Did you mean: 

convert to real world measurement

Hi Lenox,
 
The machine vision tools do not work on a binary image.  If the circles are at random locations, you can threshold the image to obtain a binary image.  Then use Particle Analysis to find information such as bounding rectangle width and height to obtain two different diameter measurements for each particle, and the elongation factor and Heywood circularity factor to find circularity information about the particles.  You may need to use some of the binary morphology tools to remove any unwanted particles that are not the circles of interest.
 
If the circles are not random in placement but occur in the same relative location to each other, then you can use pattern matching from the machine vision tools to find a frame of reference to adjust the image position.  Then you can use Find Circular Edge to find circles that occur in the same relative location.  Remember that these tools will not work on a binary image.
 
 
0 Kudos
Message 11 of 12
(1,066 Views)
hi
 
thank you .. it really helped...
one ques though..
my circle pixel data has two clusters in it.. but only the first cluster of  real world coordinates seems right.
 
my second pixel data gives x=557 y=478 r=32 and ca=4. but my real world data gives x=y=r=ca=0...
 
 
please help..
0 Kudos
Message 12 of 12
(1,108 Views)