2nd Attempt - NGC 6888

[peonic]

I managed to get my new rig out for a second stab at imaging since my return to astrophotograophy after a 10 year hiatus last night. Quite a bit of frustration getting polar alligned - but when SharpCap tells you the issue is rotation, maybe the wisest thing would be to check the camera is firmly secured first...... 😕

I got there in the end and managed to get 2 hours of subs on The Crescent Nebula.  I *THINK* the data is pretty decent given the equipment I'm using - but I certainly need to work on my processing.

 

42x180s subs using the Optolong L-Extreme filter on a ZWO ASI 183MC, HEQ5 Pro, SW ED80

 

Any processing pointers would be HUGELY appreciated. I've attached the FITS file from DSS if anyone wants to have a poke and show me where I've gone wrong!  I did try and follow a dual narrowband processing tutorial - but the resultant image was a totally hot mess!

 

Chris

DSS Stacked 16bit.FTS

Edited June 2, 2022 by peonic

[alan potts]

I think that has come out rather nicely.

Alan

[Budgie1]

You've got a lot more good data in there, Chris.

Try to stretch it a bit more and make sure you don't cut off the black by trying to make the background too dark during the stretch. You can tone this down afterwards.

You don't say what software you're using but I put your stack through my normal work flow in PixInsight and came up with the image below. I didn't want to remove the purple on the right of the image in case it was a star, so I left it in there. You can see the extra amount of detail in the Crescent and the extra background nebulosity there is in there.   

[peonic]

Holy Moly, Martin! Are you sure you used my stack?

(actually, given the whacking patch of amp glow that hasn't completely calibrated out, yes, that's my camera!)

 

For this image, I used a combination of Photoshop and Siril - but a lot of my processing techniques are those that I used years back, when shooting with a Canon 300d.

I have a trial copy of PixInsight, and I'd love to learn more about your processing workflow, if you'd care to share?

Do I basically have to start by pushing the stretch much further?

 

Chris

[peonic]

Replying to myself here.....

Just spent a few hours in PixInsight following a REALLY nice tutorial from Cuiv the Lazy Geek ....

 

Reprocessed the data as HOO, and I have to say I'm pretty pleased with the result. Can't wait for some more clear skies to hopefully add to the data.

 

Still very interested in comments, tips and techniques.  I have to say, PixInsights interface takes some getting used to - but wow that's some powerful software!

 

[Budgie1]

1 hour ago, peonic said:

I have a trial copy of PixInsight, and I'd love to learn more about your processing workflow, if you'd care to share?

Okay, it's a bit hard to explain the workflow but I'll try.

1. Now I know that's amp-glow then I put the image through GraXpert to remove the gradient left from the amp glow.
2. The open the image in PixInsight and perform a Dynamic Crop to remove any residuals from around the edges caused by stacking. Yours didn't need much here.
3. For ease, I used Automatic Background Extraction (ABE) - set to Subtract.
4. Now find a nice dark and starless area of the background and, using New Preview (Alt + N), put a box around that area. Now use this box as the reference image for Background Neutralisation.
5. Now Bin by x2 using IntegraResample tool - you don't need the full sized image so you this will tighten things up and save the computer working so hard.
6. The last thing to do before stretching is to reduce the background noise. For this I use NoiseXTerminator set at 0.7 and 0.15 for detail.
7. Now use the Historgram Transformation process to stretch the image. I only used this because the STF was too harsh on this image. I brought the midtone slider left until the image started to get lighter and then zoomed in on the historgram. This allow you to move the black point slider right until you see the number start to go up in the grey box next to the "Shadows" total. Once these numbers start to rise, move the slider back to the left until it just reads zero again, then you're not clipping the black point of the image. Continue to adjust the midtone slider until it's where you're happy with the image. 
8. At this stage I remove the stars using StarXTerminator with "Generate Star Image" ticked so it creates a new image with just the stars in it and leaves the original with a starless background.
9. Now I work on the stars - I use the Curves - Saturation to increase the colour to where I want it and then run Script > Utilities > CorrectMagentaStars. This gives a more natural look to the stars when using a duel band filter with a OSC camera. 
10. On to the main image. Use Curves - Saturation again on this one to bring the colours up to where you want them, I use RGB & L in Curves with tweak the colours and brightness. I also created a mask around the Crescent to allow me to work on the target & background separately, just by inverting the mask.
11. The final steps are to use Unsharpenmask to sharpen the image if you want too.
12. Now use PixelMath to add the stars back in. The formula is simply DSS_Stacked_16bit+Stars, tick the "Create new image" box and execute.  

I think that's about it, there is a lot of tweaking to get it to where you're happy with the image and I often find that I have to delete and come back to it, if it's not going right. Take your time, have a play and see what you can do.

A good tutorial set which I used to start off is THIS ONE. Some of the processes have been updated since  Mitch put the videos together, but the basics are there.

[peonic]

That's awesome Martin - thanks so much for your help!

[peonic]

Been trying to teach myself (AKA learn how to overcome the interface) of PixInsight - thanks to some great tutorials, and picking up hints and tips from members here I think I'm starting to get my head around it - so here's a rework of this data done solely with PI.

 

I think it looks better - but I'd be happy to hear comments on tips of things I could have done better.

 

 

[Paul Holdsworth]

Hi Chris, just had a quick look at your data and done some processing in pixinsight. There is a bit of the blue shell starting to come through from what I can see . I've rotated it but that's just a personal preference. The work flow I used was 

ABE, SCNR, Colour calibration, Noise x terminator, EZ soft stretch, Star x terminator, Star reduction (morphological) Colour mask (red) , GAME script which is the bit that picked up the blue shell, Game script mask , curves  and put the stars back. Hope this is of some use.  I'll try find the you tube tutorials for the game script and send the to you if you like ? 

Cheers

Paul

[peonic]

Links to tutorials would be awesome Paul, thank you! I've not come across the GAME scripts before!

 

Chris

[Paul Holdsworth]

Here you go Chris.  I should have mentioned I would normally add a bit of convolution to the game script mask to take away any sharp edges which I didn't do on your data but I would do that to hide where the mask has been. 

 

https://youtu.be/_ddHNpCIKo8

Hope this is of some use

Regards

Paul