Help my brain cells : Planetary camera considerations

[Altocumulus]

4 minutes ago, vlaiv said:

Correct focus position does not depend on either of those two things and I don't really see why would either over or under sampling prevent anyone from achieving good focus.

It's a bit like saying that finding correct focus for visual needs very high magnification. I've never had issues where I could not focus at particular magnification. I did however have situations where seeing was so poor that I was not certain if I had correct focus.

Thank you for the clarification.

I'll put it down to seeing conditions, which means I ought to pay more attention to the atmospheric conditions - that I do.

[chiltonstar]

If it's of any use, in practical measurements my 180 Mak with a camera with 3.75 micron pixels does show more detail at f22 (6x pixel size than f15 (4x pixel size) if the seeing is excellent.

Chris

[vlaiv]

4 minutes ago, chiltonstar said:

If it's of any use, in practical measurements my 180 Mak with a camera with 3.75 micron pixels does show more detail at f22 (6x pixel size than f15 (4x pixel size) if the seeing is excellent.

Chris

I would very much like to see that if at all possible.

 

[inFINNity Deck]

7 hours ago, geoflewis said:

Thanks for the further explanation Nicolàs, so are you saying it is the large black border in the FFT  of my Mars images that tells you it is oversampled?

Regarding using a B-Mask for planetary focusing, for me that is a complete no. Even with aftermarket mirror locks on my C14, moving the scope from a bright star back to the planet will likely see the mirror move enough to lose focus, but in any case I don't have a B-mask for the C14..... I did have one for my old 10" Meade LX200, but that never worked well enough for me for planetary imaging, so I almost always use planetary surface features to focus.

Indeed the black border tells us it is undersampled oversampled (the spectrum 'begins' in the centre). The white-noise example shows that when oversampling by a factor of two, the black border is about 50% of the frequency spectrum. So the other way around thus tells us how far the image is oversampled. Here is what seeing does (from my Dutch forum article, based on the English ArXiv article😞

The sunspot-images were generated using software I wrote for the purpose and shows how they deteriorate as a result of increasing seeing. From left to right seeing is 0", 1" and 2". The simulation was done for a C11 and ASI174MM. Sadly enough the base image was already oversampled, something we found out slightly too late. Still the effect of increasing seeing clearly shows as the central blob in the frequency spectrum contracts almost proportionally. The one of the far right is quite similar to the Jupiter image of 20 December.

Nicolàs

Edited December 23, 2022 by inFINNity Deck
Accidentally used undersampled where it should read oversampled.

[inFINNity Deck]

...

Edited December 23, 2022 by inFINNity Deck
accidental double post, som removed it

[geoflewis]

7 minutes ago, inFINNity Deck said:

Indeed the black border tells us it is undersampled (the spectrum 'begins' in the centre)

Did you mean undersampled or oversampled? I think you were saying that my Mars images were oversampled.....🤔

[inFINNity Deck]

5 hours ago, geoflewis said:

Did you mean undersampled or oversampled? I think you were saying that my Mars images were oversampled.....🤔

That should indeed be oversampled, sorry for the confusion...

Nicolàs

[inFINNity Deck]

Over the past few days I have been editing Part 2 of my Dutch article on the optimum focal ratio. I have explained that the difference between mono and colour cameras only affects the stack-size when we want to achieve the same S/N-ratio and that the effective pixel-size no longer affects the resolution (i.e. the correction factor of 2 is no longer required when stacking).

I added a section in which I show how seeing affects oversampling. As the original is in Dutch, I thought it good to provide an English version here (but Chrome does translate above link pretty well).

First we need to realise how seeing determines whether we are diffraction-limited or seeing-limited:

I have done the calculations for red, green and blue. For green I used the same wavelength (574nm) as in all the simulations in the article Van der Werf and I wrote about the influence of seeing on detail level. The central value for for instance ZWO green-filter is slightly lower, around 540nm, which would position the green graph right in between the red and blue in above figure.

So a scope with 180mm aperture (e.g. a 180mm f/15 Maksutov) would be seeing limited in green light when seeing goes above 0.8 arc-seconds. A SCT with an aperture of 279mm (C11) would be seeing-limited above a seeing of 0.5 arc-seconds.

I determined how seeing affects oversampling by running the animations I made for the article on the influence of seeing on detail level through ImageJ. The resulting frequency spectra were used to measure the oversampling manually:

On average for these two scopes the seeing affects oversampling by approximately 1.6 x [seeing] ^ 0.74.

On an ideal evening, seeing can come close to the scope's critical seeing level (i.e. close to being diffraction limited). At a best seeing of 0.6 arc-seconds, the oversampling factor will only be about 1.1, at 0.7 arc second it will be about 1.2, so the ideal f-number would need to be adjusted by a few dozen percent (slightly more with the SCT than with the Maksutov).

Last night I used my C11 EdgeHD to image Mars with my default planetary camera ZWO ASI174MM (5.9 micron pixel size) and with a recently acquired ASI290MM (2.9 micron pixel size, it was first light for this camera). So theoretically the camera would respectively require a focal ratio of 3.7 x 5.9 = 21.83 and 3.7 x 2.9 = 10.73. If we assume that we would need correct the focal ratios by about 20% due to seeing, these focal ratios would become 17.5 and 8.6 (or 19.6 and 9.7 when we correct by 10%). The C11 is f/10 and I added a TeleVue 2x PowerMate, by which it became f/20. So according to the theory the ASI290MM images would be oversampled by a factor of 2, the ASI174MM would sample correctly. The following luminance images of Mars are from around 7pm UTC:

 

A few hours later the gaps in the cloud cover became large enough for complete LRGB runs:

In both cases the image at the far right was made with the ASI174MM and resized 200% in post-processing to match the ASI290MM images. The rest of the processing was completely identical. Stacking was done based on 50% quality level in AutoStakkert!3 (with a maximum of 50% stack length). Sharpening was done during stacking with a 30% raw blend. No further deconvolution or sharpening has been applied. The only other correction (equally applied to all images) was a bit of additional saturation and correction for green.

In the monochrome images we see details in the 290MM image that we do not see that clearly in the 174MM image and vice versa. The same accounts for the colour images. As expected the 290MM images were oversampled by a factor of 2 according to ImageJ  (the ASI174MM images were only slightly oversampled, indicating that seeing was affecting imaging).

Nicolàs

 

[Altocumulus]

Thank you Nicolàs....I think!

I'll be honest, it's getting a bit too complicated! 

 

 

 

 

[geoflewis]

Thanks Nicolàs,

I remain confused. In your summary you state.....

On 27/12/2022 at 13:49, inFINNity Deck said:

In the monochrome images we see details in the 290MM image that we do not see that clearly in the 174MM image and vice versa. The same accounts for the colour images. As expected the 290MM images were oversampled by a factor of 2 according to ImageJ 

.....so the 290MM is significantly oversampled, which you think is wrong, but it produces a more detailed image....🤷‍♂️ If the images on the right (mono and colour) are with the oversampled 290MM, then IMHO they are clearly superior images. So oversampling is good, yes...?

[inFINNity Deck]

Hi Geof,

the images on the right are with the ASI174MM images, so the ones that were not oversampled during capture (but they are oversampled now because they were 200% resized).

I do not say it is wrong to oversample, but it is unnecessary and it comes at a cost: longer exposure times (or higher gain), as correctly sampling means shorter exposure times (or lower gain).

Nicolàs

[geoflewis]

37 minutes ago, inFINNity Deck said:

the images on the right are with the ASI174MM images, so the ones that were not oversampled during capture (but they are oversampled now because they were 200% resized).

I told you I was confused 😖, thanks for clarifying....😀