RE: Focus indication

[Tom Droege <tdroege@veriomail.com>]

Rob,

There are still more things you want to do with the x and y fwhm.  Again, 
the experts probably know other ways to do these things.

As I am testing MIKE, the stars are far from symmetrical points.  They are 
little streaks going from NE to SW.  This is because MIKE is set for 
Cincinnati and I am in Chicago.  Further the cart it is on is just sort of 
pointing EW.  One would like to look at the images and make some decision 
about which way to rotate the cart and tilt the polar axis.

In the past, I was not able to do this very accurately as I only had the 
images and what I could do with Image Scientist to make decisions.

But it is not so easy.  When you go out and bump the mount to rotate it, it 
may also bump the focus.  So it can be a long and tedious process to get 
things lined up.

Tom Droege

At 04:35 PM 8/1/01 -0600, you wrote:

>Thanks Michael,
>
>I wasn't considering the first method since I had gone to great lengths (~15
>lines of code :-) ) to calculate the centroid.  I didn't immediately want to
>throw away those hard 10ths of a pixel.  But, I noticed on a at least one
>star, the centroid shifted more that a pixel away from my max value, so
>assuming that was correct, the centroid is still useful.
>
>Would there be any intrinsic value in presenting FWHMx and FWHMy?
>Presumably, there would only be a significant difference during scope setup,
>when it might not be aligned.  I could possibly use the FWHM x,y to filter
>out bad/non stars, provided the majority were symmetrical.
>
>Thanks,
>Rob
>
> > -----Original Message-----
> > From: Stupendous Man [mailto:richmond@a188-l009.rit.edu]
> > Sent: Wednesday, August 01, 2001 3:40 PM
> > To: Creager, Robert S; tass@listserv.wwa.com
> > Cc: mwrsps@rit.edu
> > Subject: RE: Focus indication
> >
> >
> >
> >   Rob asked:
> >
> > > I have the centroid, but am unsure on the interpolation
> > from the centroid to
> > > calculate the FWHM.  Theoretically, I believe I should fit
> > a gaussian to the
> > > marginal sums (x and y, from star pixels only).  Is this
> > the only way, the
> > > wrong way, or is there a simpler way?  Strongly looking for
> > the simpler
> > > way...
> >
> >   Simple way:
> >
> >          - round centroid to integer (row, col)
> >          - define "peak value" as the value of pixel at (row, col)
> >                      (after subtracting background)
> >          - walk down rows in the negative direction until you reach
> >                      a pixel with half the peak.
> >                      Call this distance in pixels "d1"
> >          - walk up rows in the positive direction until you reach
> >                      a pixel with half the peak
> >                      Call this distance in pixels "d2"
> >          - walk down cols in the negative direction until you reach
> >                      a pixel with half the peak.
> >                      Call this distance in pixels "d3"
> >          - walk up cols in the positive direction until you reach
> >                      a pixel with half the peak
> >                      Call this distance in pixels "d4"
> >
> >          - calculate
> >                      FWHM = (d1 + d2 + d3 + d4) / 2
> >
> >   Complicated way:
> >
> >          - form the marginal sums in the row and column direction
> >          - fit a gaussian to the 1-D marginal sum in row direction
> >                      Set "d1" = 2.35*sigma of gaussian
> >          - fit a gaussian to the 1-D marginal sum in col direction
> >                      Set "d2" = 2.35*sigma of gaussian
> >
> >          - calculate
> >                      FWHM = (d1 + d2) / 2
> >
> >
> >   The first way is really easy and really fast.  Don't forget to
> > subtract the background from all pixels before starting, though.
> >
> >                                                 Michael Richmond
> >