How the hell do I image comets?

[MylesGibson]

I've been having loads of trouble trying to image the recent comet (C/2022 E3 ZTF). I can't seem to figure out how to process it after imaging and I fear I may have now lost my window after being sick for a while, then finally getting clear skies last night. I took some images but cannot seem to get anything from it and I think I may have wasted the opportunity.

 

I started taking a bunch of low exposure images (30s) to make sure neither the stars nor the comet were elongated. I tried the comet blend mode stacking in DSS, and it just doesn't work for me. It never gives even a slightly good result for some reason. So I stacked normally, and got a horrible streak for the comet (expected) then stacked with the comet and the stars trailed (expected) but I just couldn't get them to blend nicely in photoshop.

 

I saw that some people would take long exposures and guide on the comet to get better detail from it, so I did that. I guided on the comet for 2 mins, took a bunch of those, then did the same but guiding on the stars. Once again, stacking normally I have the streak from the comet I just cannot get rid of, and the comet images won't even stack at all. 

 

I have no idea where to go from here. Is there a technique for this or have I just wasted the night?

[happy-kat]

Hi. My prior use of DSS for stacking comets was a bit iffy with streaks but I got a bunch of data last night and I'll try the mark comet on first and last frame and then on the chosen reference frame to see if that helps.

With neowise I had better results using comet stacking in Siril.

 

[MylesGibson]

17 minutes ago, happy-kat said:

Hi. My prior use of DSS for stacking comets was a bit iffy with streaks but I got a bunch of data last night and I'll try the mark comet on first and last frame and then on the chosen reference frame to see if that helps.

With neowise I had better results using comet stacking in Siril.

 

I've never used Siril. What are the techniques with stacking comets in Siril then?

[happy-kat]

This thread might help, I'll be refreshing my memory with it later

https://stargazerslounge.com/topic/405887-comet-stacking-on-siril/

[Elp]

DSS is pretty much the same.

1. Once your images are loaded, check all and register them only (no stacking).

2. Go to settings and under the comet tab choose which type of stacking you want, comet only, stars only or comet and stars.

3. In the main window there's a button on the RHS out of the four there, it's a green star with trail (comet). Select your first image, press the green comet button and you have to manually select where your comet head is, it helps to zoom in with your middle mouse button, hold the shift key on your keyboard to be more accurate in the placing.

4. Not sure about the next step, when I did it I did the above for all frames, but I think alternatively you then select a middle/intermediate frame and highlight the comet again, then again for the last frame.

5. Stack with aggressive sigma/rejection factor like 1 and 2 otherwise your comet only stack will show harsher dots for star trails.

6. Most of the work then comes from post processing as they'll usually be artifacts to remove.

Siril is very similar, but you have more control on the stages so can apply flats, darks etc and do background extraction.

[MylesGibson]

24 minutes ago, Elp said:

DSS is pretty much the same.

1. Once your images are loaded, check all and register them only (no stacking).

2. Go to settings and under the comet tab choose which type of stacking you want, comet only, stars only or comet and stars.

3. In the main window there's a button on the RHS out of the four there, it's a green star with trail (comet). Select your first image, press the green comet button and you have to manually select where your comet head is, it helps to zoom in with your middle mouse button, hold the shift key on your keyboard to be more accurate in the placing.

4. Not sure about the next step, when I did it I did the above for all frames, but I think alternatively you then select a middle/intermediate frame and highlight the comet again, then again for the last frame.

5. Stack with aggressive sigma/rejection factor like 1 and 2 otherwise your comet only stack will show harsher dots for star trails.

6. Most of the work then comes from post processing as they'll usually be artifacts to remove.

Siril is very similar, but you have more control on the stages so can apply flats, darks etc and do background extraction.

This is what I tried originally but it just comes out terrible for me!

[Elp]

Was you tracking in sidereal? When I did it for an hour the background stars basically didn't move from start to finish.

What do your stars and comet only images look like?

[MylesGibson]

33 minutes ago, Elp said:

Was you tracking in sidereal? When I did it for an hour the background stars basically didn't move from start to finish.

What do your stars and comet only images look like?

You tracked the comet for an hour without the background stars moving??? I was guiding, I did some where I guided on the stars, and some where I guided on the comet

[Elp]

I was surprised too as most of the time I autoguide. Please post the pics here, I'm suspecting your comet is moving around a lot. I think a lot of these stacking solutions work via the comet moving in one vector only in a linear fashion, it won't stack well if it's moving around a lot in the frames.

You will have to clean up your images afterward, I probably took three attempts to get mine right and spent just as much time on it as I do DSO images.

The easier option will be to edit a singular image but where's the fun in that?

[MylesGibson]

1 hour ago, Elp said:

I was surprised too as most of the time I autoguide. Please post the pics here, I'm suspecting your comet is moving around a lot. I think a lot of these stacking solutions work via the comet moving in one vector only in a linear fashion, it won't stack well if it's moving around a lot in the frames.

You will have to clean up your images afterward, I probably took three attempts to get mine right and spent just as much time on it as I do DSO images.

The easier option will be to edit a singular image but where's the fun in that?

I'll have to add them here once I get home from work as they are on my home laptop! The comet was moving in a linear fashion as can be seen by the star stacking and the quite straight line moving through the image where the comet was moving. It may be that the exposure time was maybe too long so when I was guiding on the stars, it did elongate the comet in a single 120s exposure. And obviously the stars trailed when doing a 120s exposure while guiding on the comet

[Varavall]

10 hours ago, MylesGibson said:

I've been having loads of trouble trying to image the recent comet (C/2022 E3 ZTF). I can't seem to figure out how to process it after imaging

Hi

I use ASTAP and do manual alignment.

 

And the result is

 

As can be seen I am using a mono camera just to make life more difficult. I am trying to create a starless comet image that I can slap on a starry background, but that is proving taxing. at the moment.

PS Some success. I used Starnet++ CLI to process all the images for each filter, then stacked and aligned in ASTAP, and combined the star field images and the comet images in GIMP and the result is

 

I think it needs a beter star field.

Good luck

Adrian

Edited January 31, 2023 by Varavall
Additional material

[Elp]

What are the settings for the stacking and rejection? They need to be quite low in order to force the rejection of the stars, you will still have some to remove manually too after the stacking especially if they run over the comets back end.

[inFINNity Deck]

Here is a nice tutorial (never tried it myself though): 

I do use APP, but not the Starnet++ option, simply allow trails to show:

Nicolàs

 

[900SL]

If you use DSS, make sure you select your best frame as the reference frame (right click on it in the file list) and highlight the comet in first, last and reference frame as noted above

 

 

[MylesGibson]

15 hours ago, inFINNity Deck said:

Here is a nice tutorial (never tried it myself though): 

 

I do have APP, so I could give this a go. Thanks!

 

15 hours ago, Varavall said:

I use ASTAP and do manual alignment.

I've not used that before. Is it free software?

 

15 hours ago, Elp said:

What are the settings for the stacking and rejection? They need to be quite low in order to force the rejection of the stars, you will still have some to remove manually too after the stacking especially if they run over the comets back end.

Good question. I'll have to check. I had the scope out again last night and completely forgot to upload the pictures I said I would! Sorry

 

11 hours ago, 900SL said:

If you use DSS, make sure you select your best frame as the reference frame (right click on it in the file list) and highlight the comet in first, last and reference frame as noted above

I did do that, but still came out really weird! Will see what happens when I put some more time on it tonight!

[carastro]

I haven’t done a comet for a while but l found DSS comet stacking to be very temperamental.  Occasionally it would work but mostly it didn’t. 

Think l ended up stacking manually in the end in photoshop.

I  think these days there are many more software options around and options for star removal that l would be stacking twice, once on the comet  and then remove the stars and then a separate stack aligning the stars.  Then select the stars and place them in the comet only image. 

carole

 

[TheDrunkAstronomer]

I can't get much more than a green blur out of it.

[MylesGibson]

38 minutes ago, carastro said:

I haven’t done a comet for a while but l found DSS comet stacking to be very temperamental.  Occasionally it would work but mostly it didn’t. 

Think l ended up stacking manually in the end in photoshop.

I  think these days there are many more software options around and options for star removal that l would be stacking twice, once on the comet  and then remove the stars and then a separate stack aligning the stars.  Then select the stars and place them in the comet only image. 

carole

 

I think this may be the way to go, although I did try stacking just the comet but even that didn't turn out great. I might try something other than DSS as I have never had much luck with the comet stacking!

[inFINNity Deck]

3 hours ago, MylesGibson said:

18 hours ago, inFINNity Deck said:

Here is a nice tutorial (never tried it myself though): 

 

I do have APP, so I could give this a go. Thanks!

Registering and stacking on a comet is a breeze using APP (once you found how to do it): https://www.astropixelprocessor.com/community/comet-registration-and-integration/comet-registration-in-astro-pixel-processor-1-083-beta1-video-tutorial/

The latest APP has star removal tool as well. Later this week I hope to find some time to test the star removal option in APP to see if it is possible to create a comet image with round stars using only APP and image processing software (I use PaintShop Pro).

Nicolàs

PS: the latest APP version has a lay-out bug as a result of which the comet nucleus selecting buttons disappear off-screen when using a pc with low screen resolution. Version 1.083 does not show this behaviour, but does not have the remove star option. I have reported this bug and hope Mabula produces a fix soon.

[MylesGibson]

30 minutes ago, inFINNity Deck said:

Registering and stacking on a comet is a breeze using APP (once you found how to do it): https://www.astropixelprocessor.com/community/comet-registration-and-integration/comet-registration-in-astro-pixel-processor-1-083-beta1-video-tutorial/

The latest APP has star removal tool as well. Later this week I hope to find some time to test the star removal option in APP to see if it is possible to create a comet image with round stars using only APP and image processing software (I use PaintShop Pro).

Nicolàs

PS: the latest APP version has a lay-out bug as a result of which the comet nucleus selecting buttons disappear off-screen when using a pc with low screen resolution. Version 1.083 does not show this behaviour, but does not have the remove star option. I have reported this bug and hope Mabula produces a fix soon.

Oh fantastic! I'll have a go at that tonight!

[happy-kat]

I currently trying ASPAP, have 317 20s images

[inFINNity Deck]

Today I gave APP 2.0.0 beta 13 a try. What I did was as follows:

– First stack on the comet, so with Registration Mode on Comet/One Star, and with Integration set to a Median Filter with Kappa High to approximately 1.5. This produces a sharp comet with no stars. I will call this [Plate 1].
– [Plate 1] still shows some vague traces, but these can easily be removed with a clone brush. It took me about 5 minutes to clean [Plate 1] a bit further.
– Then stack on the stars with default settings, resulting in a starry sky with a smeared comet. This is [Plate 2].
– Pass [Plate 2] through APP's Star Removal Tool, which results in a star-less smeared comet. This is [Plate 3].
– Ran all three plates through Topaz to reduce the noise, but above all to reduce the artefacts left over from the Star Removal Tool, otherwise they will reappear later when combining.
– Load [Plate 3] as a new layer on top of [Plate 2] and set this layer to difference. With this we subtract the smeared comet from [Plate 2], leaving a clean starry sky. This is [Plate 4].
– Load [Plate 1] as a new layer on top of [Plate 4] and set this layer on screen, so that the two layers are combined. This is [Plate 5].
– Then edit [Plate 5] as a normal deep-sky image until it is to your liking.

The end result is what you see here:

When enlarged it becomes visible that I followed the comet and used a too long shutter speed, so the stars are slightly oval. This can be prevented by using shorter exposures and more subs. The background also shows some streaks in the direction of the comet's orbit. This could be reduced with some extra cleaning of [Plate 1].

The question is, of course, whether this is a good representation of reality. The integration took almost an hour and the comet is now approximately where it was halfway through that period, but whether this is actually correct is questionable. For this plate one might as well combine [Plate 1] with an image of the same part of the sky taken at a later date, then the stars can be much rounder.

Nicolàs