From: "Dwight K. Elvey" 
To: atm{at}shore.net
Reply-To: "Dwight K. Elvey" 


Hi
 I forgot to mention, I believe that you can
take the data from the pixels of a number of lateral positions and then run
what is called an "auto-correlation"  on it. It should make peaks
where both sides match best. The two input streams will be from the left
side of the mirror to the center and then from the center to the right
side. You take the pixels over time as the lateral motion.
 I think this is right anyway.
Dwight

>From: "Dwight K. Elvey" 
>
>>From: "James Lerch" 
>
>---snip---
>
>Hi James
> When I measure by eye, I don't set a predetermined
>intensity. What I look for is that a particular zone
>grays at the same rate from both sides as I move the
>edge laterally. This seems different than what your
>code is doing.
> Perhaps you can do the same by measuring several pictures
>as you move the lateral position of the KE and look to
>see what zones on both sides change at the same rate.
>You should find that on either side of the correct zone,
>the light will be stagered when the intensity changes.
>One side will lead and the other will lag. The two
>sides don't even have to be the same exact intensity
>since errors in the camera and light source contribute
>to this while the change rate is only related to
>the zones distance from KE. You can even normalize
>the curves from total illumination to total dark to
>compensate for any errors.
> I realize that this is quite different than what you
>are doing but it is the only way that I've been able
>to finely pin down Foucault readings.
>Dwight
>
>Some ascii art follows:
>
>Light curves for zones with lateral KE motion:
>
>Zones not matched:
>
>   ____
>R      \_____
>   _____
>L       \____
>
>
>Zones matched:
>
>   ____
>R      \_____
>   ____
>L      \_____
>
>
>
>
>
>

--- BBBS/NT v4.00 MP