Re: ATM An efficint algorithm for robotic Foucault testing

["Bob May" <bobmay@nethere.com>]

When you do the analysis of the data from the image, you want to find the
vertical limits as well as the horizontal limits of the image.  If the
mirror moves across the image, you have a problem if you do the centering
once so be careful in that arena.  After that, you have to insure that the
bright areas aren't saturated nor black nor too far from those limits.  This
keeps the brightness of the light coming back in the linear region of all
the electronics between the light and the pixel values in the memory of the
computer.  You then take and find the average between and use that value to
determine the position of the KE across the light beam and then compare the
brightness from side to side for the "correctness" of the position of the KE
to the mirror.
At that point, you should be able to determine the accuracy of the surface
you are measuring to a parabolic surface.
Bob May
http://nav.to/bobmay
bobmay@nethere.com
NEW! http://bobmay.astronomy.net