Star Count differential

[mohic2020]

Hi,
To avoid false positives, reflections, and other light artifacts. Couldn't we make a difference between two images stars based on two temporally shifted images?
In the same way that we do treatments to eliminate hot pixels.

Clear Sky

[EricClaeys]

How exactly would this work, given that the camera doesn't move? The stars are in slightly different places each picture.

[mohic2020]

yes, precisely, only the stars move during the Earth's rotation. the difference between two images helps eliminate reflections

[aaronwmorris]

Unfortunately, it is not as simple as you would think it would be. It is not a matter of just rotating the image. The fisheye distortion from an allsky lens causes object to not move in the path of a perfect circle. The closer you get to the edge of the image, the greater the effects of the distortion.

[mohic2020]

yes, but I already use a mask to only take the upper part of the sky, around 30° around the zenith. There is little or very little distortion there, right?

[EricClaeys]

I suppose that would depend on your dome, lens, and how perpendicular your camera/lens is to the dome.

[mohic2020]

I did the basics.
dome of 5cm radius.
lens 2.5mm F1.6 at 2cm from the dome, which gives the two parallel curvatures.