#1  
Old 15-01-2020, 01:59 PM
Paulyman (Paul)
Registered User

Paulyman is offline
 
Join Date: Nov 2017
Location: Unanderra
Posts: 130
One Year of Processing Progress

Hi all,

Since the smoke is still hanging around, as well as the much needed showers, I thought I'd go back and look at my old data.

I decided to have a go at reprocessing my first DSO image of M42 taken back in October 2018. I used my WO Star 71 and Canon 550D on the HEQ5 before I had worked out guiding. The image consisted of 5 hours of 60s subs, unguided and with the camera sensor reading 30C according to BYEOS.

At the time all I really knew how to do was use levels and curves in photoshop, I had no idea about noise reduction or masks etc. The first image below was the best I could achieve at the time.

Yesterday I decided to have a crack at reprocessing the image from the ground up using PixInsight. Now it is far from perfect, I couldn't save the blown out core for instance but I assume I'd need some much shorter subs and some HDR for that. It is much better than the original attempt though, look at all that dust that was just hiding in the noise!
Attached Thumbnails
Click for full-size image (original_overstretched.jpg)
203.8 KB77 views
Click for full-size image (final_02_nohdr.jpg)
149.7 KB71 views
Reply With Quote
  #2  
Old 15-01-2020, 02:02 PM
LewisM's Avatar
LewisM
Novichok test rabbit

LewisM is offline
 
Join Date: Aug 2012
Location: Somewhere in the cosmos...
Posts: 9,442
Sorry, but the original is more representative of the actual colours.

You'll need either more data, or to blend / layer in differently stretched versions.

Left is perfectly fine for the amount of data, whereas the right is pushed way too far.
Reply With Quote
  #3  
Old 15-01-2020, 04:36 PM
Atmos's Avatar
Atmos (Colin)
Ultimate Noob

Atmos is online now
 
Join Date: Aug 2011
Location: Melbourne
Posts: 6,392
Left is definitely my favourite of the two

It’s a very natural representation
Reply With Quote
  #4  
Old 15-01-2020, 05:33 PM
Paulyman (Paul)
Registered User

Paulyman is offline
 
Join Date: Nov 2017
Location: Unanderra
Posts: 130
Okay, so it appears that when I saved it as a tiff to open in photoshop I left it at the default 32 bit, which in my PS at least causes the image to blow out the highlights and saturation when opened, not sure why. I didn't notice I just opened, resized and saved. So that image got posted to astrobin and instagram as well, oh well. On a phone it doesn't look so bad. I've just found the resample process in PixInsight so hopefully that won't happen again.

But since I was resaving it anyway, I took the advice above and massively toned back my contrast adjustments, mostly the saturation.
Attached Thumbnails
Click for full-size image (minimal.jpg)
166.5 KB44 views
Reply With Quote
  #5  
Old 15-01-2020, 06:18 PM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 357
Hi Paul,
A great image with sharp stars and good data.
Hope you don't mind but I just shifted the black point a bit and a slight curve adjustment in PS. but thats my personal preference.
If you have some shorter subs you can use a mask in the core area to balance it out a bit but other than that its a really good image.
Cheers
Andy
Attached Thumbnails
Click for full-size image (Paulymans image.jpg)
192.9 KB36 views
Reply With Quote
  #6  
Old 15-01-2020, 09:00 PM
Atmos's Avatar
Atmos (Colin)
Ultimate Noob

Atmos is online now
 
Join Date: Aug 2011
Location: Melbourne
Posts: 6,392
It’s definitely better but I still prefer the original myself. It looks more natural.
I think it’s what the HDRTransformation tool does, I’ve had the same issue with my own versions of M42. I’ve been playing around with PixelMath trying to solve the issue and I’m getting close to a fix which involves some mathematical masking between a more heavily stretched image and another that’s not overly stretched but shows the bright core well.

http://www.chart32.de/images/objects...M42/M42_80.jpg

The link above shows what I’m trying to achieve, it retains accurate colour and fine contrast. It looks flat but not artificially
Reply With Quote
  #7  
Old 16-01-2020, 12:29 PM
iborg's Avatar
iborg (Philip)
Registered User

iborg is offline
 
Join Date: Feb 2015
Location: Lynbrook, Australia
Posts: 322
Hi


I also prefer the look of the original, but, I cab see what you mean about the dust hiding in the noise.


Philip
Reply With Quote
  #8  
Old 17-01-2020, 04:09 PM
LostInSp_ce's Avatar
LostInSp_ce
Unregistered User

LostInSp_ce is offline
 
Join Date: Mar 2017
Posts: 451
As the others have expressed your original M42 has a more natural look but I do like how you've been able to eliminate the the noise and reveal the surrounding dust in your updated version. It's a good thing to process old data as we progress because we learn many new tricks along the way and software improves as well which can really make a difference. Nice work on producing a cleaner result.
Reply With Quote
  #9  
Old 18-01-2020, 03:14 PM
Cimitar (Evan)
Evan Morris

Cimitar is offline
 
Join Date: Dec 2006
Location: Gunnedah, NSW
Posts: 250
Quote:
Originally Posted by Paulyman View Post
Okay, so it appears that when I saved it as a tiff to open in photoshop I left it at the default 32 bit, which in my PS at least causes the image to blow out the highlights and saturation when opened, not sure why. I didn't notice I just opened, resized and saved. So that image got posted to astrobin and instagram as well, oh well. On a phone it doesn't look so bad. I've just found the resample process in PixInsight so hopefully that won't happen again.

But since I was resaving it anyway, I took the advice above and massively toned back my contrast adjustments, mostly the saturation.
Hi Paul

I think your new image looks great
The difference in signal to noise between both images is certainly noticeable and you've really managed to draw out those fine wisps of gas in your most recent attempt.

Well done, I'd say the past 12 months have definitely been worth it.

Cheers
Evan
Reply With Quote
  #10  
Old 18-01-2020, 04:19 PM
Paulyman (Paul)
Registered User

Paulyman is offline
 
Join Date: Nov 2017
Location: Unanderra
Posts: 130
Thanks for the comments all. So it appears my noise reduction skills have improved but my skills with colour management have not. So gives me an area to work on. More data will always help but who knows when in the near future that will happen.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 11:05 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
Advertisement
Astromechanics
Advertisement
Bintel
Advertisement
Testar
Advertisement
EQ8-R
Advertisement
Star Adventurer
Advertisement
Celestron RASA
Advertisement
Lunatico Astronomical
Advertisement
Astronomy and Electronics Centre
Advertisement