Auto find star should be more robust

[GoogleCodeExporter]

I was imaging a target that had a bright star in the field, so I was using very 
short exposures to get a high frame rate.

After SGP switched targets, there was not as bright a star in the field, and so 
because of the short expsoure time, PHD failed to lock and the sequence aborted.

Perhaps auto find star should have the ability to increase the exposure if it 
needs to. Or it should always start at some user specified value, and then 
reduce it while the star mass is above a specified value.

Original issue reported on code.google.com by br...@boneheads.us on 29 Mar 2014 at 2:40

[GoogleCodeExporter]

Original comment by andy.gal...@gmail.com on 2 Apr 2014 at 3:56

• Changed state: Accepted
• Added labels: OpSys-All
• Removed labels: OpSys-

[GoogleCodeExporter]

Here's what I'm thinking of doing:
  Add an "Auto" selection to the exposure duration drop-down list.
  In the brain (camera tab?) add a group "Auto Exposure Settings" with
     Target SNR, default 4.0
     Min Exposure, default 1.0
     Max Exposure, default 8.0 (?)

When exposure is "Auto":

Auto-find star (menu or server apis): use Max-exposure duration
  (maybe try with current frame first before switching to max-exposure?)

After each camera frame, adjust exposure duration
    assume SNR is proportional to sqrt(duration)
    new_exposure = cur_exposure * (target_snr / cur_snr)^2
        clamp new_exposure to (min_exp,max_exp) range

Original comment by andy.gal...@gmail.com on 16 May 2014 at 7:36

• Changed state: Started

[GoogleCodeExporter]

This issue was updated by revision r853.

Original comment by andy.gal...@gmail.com on 21 May 2014 at 3:26

[GoogleCodeExporter]

Issue was fixed by the Auto Exposure feature in v2.3.

Original comment by andy.gal...@gmail.com on 16 Jul 2014 at 8:51

• Changed state: Fixed