HI Robin
I was watching a interesting video from Franklin Marek (Seti Astro) where he is talking about various calibration frame types.
Youtube link
https://www.youtube.com/watch?v=3kc5qe4OwY0&t=933s
In his video , he seems to be saying , for modern CMOS cameras, the BIAS is automatically compensated for in the camera electronics during image capture.
So I would assume from this , Bias frames are no longer required for most modern imagers?
Just curious on your thoughts on this.
Take care
Steve
Bias not needed with modern CMOS cameras?
-
admin
- Site Admin
- Posts: 17558
- Joined: Sat Feb 11, 2017 3:52 pm
- Location: Vale of the White Horse, UK
- Contact:
Re: Bias not needed with modern CMOS cameras?
Hi Steve,
it's a complex question - there are a number of uses for bias frames, some of which still apply, some of which have been pointless with all cameras, and some which might have less relevance with CMOS cameras due to the reduced read noise.
In brief...
If you are just taking dark and light frames, and the light frames and the dark frames have exactly the same gain/exposure as the lights then bias frames just cancel out in the image processing - in theory you subtract the master bias from both the master dark and the light frames, but then when you dark subtract, the bias completely cancels, so you might as well have not bothered with it. There might be a minor benefit here in improving the detection of hot pixels in dark frames when the bias has been subtracted, but this can be handled statistically. Note that bias frames *do* play a useful role if for some reason you need to use dark frames that don't match your light frames (different exposure maybe)
If you are using flat frames then you need to subtract *either* bias or dark flat frames from the flat frames (and also subtract darks, or possibly bias from the light frames) to get proper flat correction.
cheers,
Robin
it's a complex question - there are a number of uses for bias frames, some of which still apply, some of which have been pointless with all cameras, and some which might have less relevance with CMOS cameras due to the reduced read noise.
In brief...
If you are just taking dark and light frames, and the light frames and the dark frames have exactly the same gain/exposure as the lights then bias frames just cancel out in the image processing - in theory you subtract the master bias from both the master dark and the light frames, but then when you dark subtract, the bias completely cancels, so you might as well have not bothered with it. There might be a minor benefit here in improving the detection of hot pixels in dark frames when the bias has been subtracted, but this can be handled statistically. Note that bias frames *do* play a useful role if for some reason you need to use dark frames that don't match your light frames (different exposure maybe)
If you are using flat frames then you need to subtract *either* bias or dark flat frames from the flat frames (and also subtract darks, or possibly bias from the light frames) to get proper flat correction.
cheers,
Robin