Automatic compensation for field rotation in alt/az mounts during high speed capture

[Borodog]

Title says it all. This will necessitate saving in a debayered format. That's ok, as videos with field rotation are typically debayered during derotation anyway. Rebayering them does not produce the best results after stacking.

[admin]

Hi,

how long are you planning to record for? Except for very close to the zenith, I'd have thought that the amount of field rotation during a typical recording of 30s to a couple of minutes is so small as to not be noticeable. If you are not in the tropics then there's nothing much of interest for high speed imaging near the zenith.

cheers,

Robin

[Borodog]

Robin,

It's not for me. I have no alt/az mounts. However, I do know that for people that do have alt/az mounts that do planetary and lunar imaging, it is absolutely a problem that has been discussed extensively on Cloudy Nights in both the planetary and lunar imaging forums. In planetary, people use WinJUPOS in post to de-field-rotate OSC video streams (ending up with a debayered video by the way; you can rebayer the video but it doesn't yield the best result). Anything over 75 seconds and it makes a noticeable difference. Lunar is much worse because of the larger fields of view. People have gone to the length of buying expensive motorized camera rotators to combat field rotation. Autostakkert has a feature to de-field-rotate, but only for mono data, not OSC. All of these (except the motorized rotators) necessitate extensive preprocessing of videos to get to a stackable result. It seems intuitive to me to just correct it before it goes in the can. SharpCap would need to know the observers latitude & longitude, and the position of the target in the sky, the former the user can enter and the latter it doesn't seem to hard to get if the user has an internet connection and sets the target. You'd need to debayer the image (I suggest the HQ Linear debayer algorithm used in PIPP; it's the best I've tested and easily beats bilinear after stacking) and apply a rotation to every frame. I'll admit, it's possible that this could slow down frame rates. This is not really a problem for lunar, but planetary imagers may not like it. I don't think there's any way to know without testing though, and it may be ROI and frame rate dependent. In other words smaller ROIs or lower (but still reasonable) frame rates may not be a problem.

[admin]

I will have to work through the calculations at some point (I have the formulae down somewhere) to see what sort of rotation rate might need to be adjusted. My gut feeling is though that this is one of those things best fixed with hardware (like it's better to use an ADC than to align the red/blue channels with the green in software - the ADC does a better job because it aligns *within* the channel as well as between them).

cheers,

Robin

[Borodog]

I take your point, but currently most people fix it in software rather than hardware. They just do it in post rather than during capture. I might consider switching to an alt/az mount, because eq mounts don’t play well with ADCs, except for field rotation. And most mounts that are sold are alt/az.