Processing challenge - M57 - Forums

Topic starter

Nov 12, 2021

I am curious, what can you do with <10 minutes of exposure time on the Ring Nebula? Trust me- this is no ordinary data set.

I have spent years sitting on this data, trying to get it just right. There are just so many artifacts due to multiple factors such as poor calibration, lack of integration time, and no anti-blooming on the chip.

I would love to see what others can do with this, I finally think I am nearing a spot where I am happy with it. I have only ever had my own eyes on this data though, so I don't really have anything to compare against.

What can you do with it? You can download the master frames from the link below

https://drive.google.com/drive/folders/1qAeW1l1IIRD0LbV7V_fpmi-HpScjRYN1?usp=sharing

Of course you can not use the above for any commercial use / submit to APOD. However, you are more than welcome to share it on your personal page. All that is asked is to include a note that the results were made using the Lowell Observatory’s Discovery Channel Telescope.

Looking forward to seeing your results!

Connor

Helpful Respectful Engaging

Connor Matherne

Topic starter

Nov 13, 2021

Here is a preview of what the image looks like roughly when combined and with the auto-stf applied. A great set of data, despite the artifacts! The halo definitely is fairly extensive, and the galaxy IC 1296 is equally great.

andrea tasselli

Frequent contributor

Nov 13, 2021

The dataset isn't great. On top of the very obvious belemishes (bleeds, hot pixel galore, saturated pixels) the use photometric filters makes it impossible to give back a natural colour feel to either stars or nebulae. This said, it is amazingly deep (well, not so amazing considering it comes from a 4+ meter scope) and the FWHM is around 1", which is kind of standard for pro-observatories of the pre-active/pre-adaptive era. Not sure it is worth spending more time in figuring out about removing the bleeds but here is my provisional result( screen shot capture):

Helpful Insightful Respectful

Andy Wray

Nov 14, 2021 · Edited: Nov 14, 2021

I couldn't get much from this data either, however I'm sure someone on here who has more than a couple of days of PixInsight experience could turn your data into quite a good image. It would be interesting to see what could be done.:

Andy Wray

Nov 14, 2021

I know I said there wasn't much in there, but with a bit of processing the eye itself has quite a bit of detail:

Connor Matherne

Topic starter

Nov 14, 2021

andrea tasselli:
The dataset isn't great. On top of the very obvious belemishes (bleeds, hot pixel galore, saturated pixels) the use photometric filters makes it impossible to give back a natural colour feel to either stars or nebulae. This said, it is amazingly deep (well, not so amazing considering it comes from a 4+ meter scope) and the FWHM is around 1", which is kind of standard for pro-observatories of the pre-active/pre-adaptive era. Not sure it is worth spending more time in figuring out about removing the bleeds but here is my provisional result( screen shot capture):

This data set definitely isn't the cleanest- but it is very deep relative to most images of the target I feel. While this data is no where near as clean as most of what I work with, that is what makes it a challenge (and interesting in my eyes!)

See below another example from this same scope I worked on (RGB / finished image). Data was rough around the edges for sure, but processing can really bring it a long way.

The M57 and above example though are both older results before I helped refine the shooting process with the operator. We eventually got to results like this

which are astronomically better and really living up to the equipment this scope has under the hood. That Needle set though is unbelievably clean, processing it is a walk in the park compared to some of these earlier sets!

Andy Wray:
I know I said there wasn't much in there, but with a bit of processing the eye itself has quite a bit of detail:

I like both of your versions Andy. What I had an issue with in processing was getting both the halo to show, and inner ring as well while also not making a ghastly transition between the two.

Connor

Helpful Insightful Respectful Engaging Supportive

IzaakC

Nov 20, 2021

Pixinsight and Photoshop

Connor Matherne

Topic starter

Nov 20, 2021

Pixinsight and Photoshop

Nice job Izaak! Thanks for taking the time to process it!

Well Written Respectful Supportive

andrea tasselli

Frequent contributor

Nov 20, 2021 · Edited: Nov 20, 2021

2nd attempt at it (actually 3rd, but this is from scratch). Changed the main processing (hysto stretching followed by log scaled autohysto stretching). Managed to return the 1st halo to a more teal-ish colour (as well as make it more prominent) and rebalanced the blue channel to have a better star colour distribution. Still not happy with the pinkish hue of the h-alpha dominated inner shell. And still no answer for dealing with blooming but might have an idea on how to deal with the bands. This for the next time though...

Sigga

Nov 20, 2021

Hi Connor,

Looks like fun to try so if ever can complete download I will.

I using Chromebook and two things apparent:

-no one at Google uses one to download from Drive, most idiotic UI ever

- Google clearly lie about having that giant data center here as download speed is 14.4 modem speed so I assume I downloading this from universe far far away

Rant complete :-)

Sigga

Michael Feigenbaum

Nov 22, 2021

M57 Processing Challenge (M57 Processing Challenge)

So a bit of an explanation... The data was really different stuff than I am used to working with so I had to try figure out where to even start! I only used the Ha, R, V, and B data and made the assumption the V was green. I believe the "U" frame was maybe ultraviolet? Even so, I had no idea what to do with it, lol...

My focus was to try to show the faint halo as well as possible along with some detail in the brighter part of the PN. For color calibrations, I did a simple calibration in PI using IC1296 as the white reference. I kind of eye-balled the color if IC1296 as a check for "reasonableness".

I left the green in the central portion of the PN as is as I believe that is close to the natural color, could be wrong on that of course.

I did a little bit of non-linear NR and a little MLT sharpening and that was really about it.

Thanks for making this available to play with Connor! It was really a lot of fun and a privilege to work with real professional data like this.

Helpful Respectful Engaging Supportive

Steve Cooper

Nov 22, 2021

Connor Matherne

Topic starter

Nov 22, 2021

Fantastic work Michael, thank you for taking a stab at the data!

Michael Feigenbaum:
So a bit of an explanation... The data was really different stuff than I am used to working with so I had to try figure out where to even start! I only used the Ha, R, V, and B data and made the assumption the V was green. I believe the "U" frame was maybe ultraviolet? Even so, I had no idea what to do with it, lol...

Good assumptions all around, and indeed those are the only frames I used. The I is infrared. Later images we ended up just using R, V, and B along with a VR filter which acted as a L replacement.

Michael Feigenbaum:
My focus was to try to show the faint halo as well as possible along with some detail in the brighter part of the PN. For color calibrations, I did a simple calibration in PI using IC1296 as the white reference. I kind of eye-balled the color if IC1296 as a check for "reasonableness".

I left the green in the central portion of the PN as is as I believe that is close to the natural color, could be wrong on that of course.

I did a little bit of non-linear NR and a little MLT sharpening and that was really about it.

Thanks for making this available to play with Connor! It was really a lot of fun and a privilege to work with real professional data like this.

I did something similar with the color calibration, however ended up masking out a lot of the green and boosting the reds and blues.

I generally go away from green in my astro images, and push it to cyan instead. This image was definitely no exception.

Personally though, I love how you accented the inner bits of the Ring, and let the outer halo be more muted and faded into the background. Shows its existence without distracting from the beauty of the inner structures.

Great job Michael, thanks again. Definitely gave me a fresh perspective on this target.

Connor

Helpful Insightful Respectful Engaging Supportive

Steve Cooper

Nov 22, 2021

I made several masks from a very exaggerated stretch of the data in PI. Then had some fun with the curves transformation tool adding selective brightness, sharpening, and color saturation with the different masks in place. Not very scientific, but a lot of fun!
Thanks

Connor Matherne

Topic starter

Nov 22, 2021

Steve Cooper:

Another great one, nice job Steve! I love the color variation in your halo, as well as how well your fixed the blooming. What did you do for the blooming? I just ended up mirroring the other half of the star and clone stamped to my hearts content.

Respectful Engaging Supportive

Michael Feigenbaum

Nov 22, 2021

Connor Matherne:
I did something similar with the color calibration, however ended up masking out a lot of the green and boosting the reds and blues.

I generally go away from green in my astro images, and push it to cyan instead. This image was definitely no exception.

Personally though, I love how you accented the inner bits of the Ring, and let the outer halo be more muted and faded into the background. Shows its existence without distracting from the beauty of the inner structures.

Great job Michael, thanks again. Definitely gave me a fresh perspective on this target.

Connor

Thank you again Connor, it really was a privilege!

Well Written Respectful Supportive

Steve Cooper

Nov 22, 2021

Connor Matherne:
Another great one, nice job Steve! I love the color variation in your halo, as well as how well your fixed the blooming. What did you do for the blooming? I just ended up mirroring the other half of the star and clone stamped to my hearts content.

Thanks Connor. I used CCDStack to remove the blooms. It helped a little bit, but I ended up using the clone stamp too