How can I accurately measure the diameter of a hole using digital photography?

[Irja]

Hello,

I am trying to measure diameter of the hole in my equipment by doing the photography with digital camera.

The result is in the picture.

If I am calculating the pixels - shall I consider only the "pure white" ones or also the grey pixels? Are the gray pixels parts of the hole or it's just the diffraction? If I shall also consider the gray ones, how shall I justify including them?

 

[mgb_phys]

First can you set the camera to save anything other than jpeg? This will help avoid the artifacts where there is a big change in brightness.
You want to measure across the 50% grey level pixels, however you do it will only be an estimate.
I would load the picture into an image package and estimate the centre of the hole, then make as many radial distance measurements to an estimate of the mid grey at different angles and average them.
You could also try setting the image to pure black/white and trying the same thing.

It probably isn't worth trying to get more complicated with automatic fitting routines with a poor image - you would be betterr off getting a microscope.

 

[Andy Resnick]

I second mgb- your 'measurement' isn't accurate enough to give you results worth anything.

That said, the method you are using (but done under very controlled circumstances) is what is used to locate the center of objects to sub-pixel resolutions. That's useful for studying molecular motors, for example. Or colloid dynamics.

I guess I should ask why you are measuring the hole diameter? About how big is the hole- microns or millimeters? There's lots of ways to measure a diameter.