Re: [ATM] Computing surface from Ronchi image?

[Jim Burrows <burrjaw@earthlink.net>]

At 2006-02-01 13:44 -0800, Dale Eason wrote:

>Is it possible to compute the actual mirror surface
>from Ronchi images similar to what you can do with
>interferometric fringes?

The Misell algorithm, a back and forth iteration of the Fourier transform, 
was used to find the Hubble mirror's deviations after it had been chucked 
up into orbit (the great demonstration of NASA's incompetence).  I tried 
the algorithm on whole-mirror images with source at the ROC, but it 
wouldn't converge.  I guess, in the Hubble case, it could be assumed that 
the mirror had only a correction error.  If a Hartmann-type mask is put on 
the mirror, the algorithm can be modified slightly and it will converge for 
general errors.  So perhaps this could be applied (without a mask) to a 
Ronchi image to cure Suiter's main objection to the Ronchi test, p. 297, 
"... [Ronchi] patterns can differ little from the patterns with 
unacceptable correction errors."

         -- Jim Burrows
         -- mailto://burrjaw@earthlink.net
         -- http://home.earthlink.net/~burrjaw
         -- Seattle N47.4723 W122.3662 (WGS84) 

_______________________________________________
ATM mailing list http://www.atmlist.net/