Jump to content

Stargazers Lounge Uses Cookies

Like most websites, SGL uses cookies in order to deliver a secure, personalised service, to provide social media functions and to analyse our traffic. Continued use of SGL indicates your acceptance of our cookie policy.

stargazine_ep29_banner.thumb.jpg.da7f3b163f7bd35187cb558b0346baf6.jpg

vlaiv

Bubble nebula - modified SHO, first NB image

Recommended Posts

3 minutes ago, RolandKol said:

@vlaiv I could not hold my self without inserting my 2p ;) tried to bring out the dark clouds on the bubble :)

a bit overcooked :) 

I should be posting raws in a few days, have been working on some improvements, so there will be a bit more SNR to play around :D

 

  • Like 2

Share this post


Link to post
Share on other sites

Fine image indeed. I like your last colour version (2nd one)

Share this post


Link to post
Share on other sites

First of all I love your image, some great data you've collected.  I've yet to process in narrowband but was intending to follow the instructions of Steve Richards in his recent book, 'Dark Art or Magic Bullet' where he talks about using the Selective colour dialogue box in PS  - for hubble palette images.  Am sure others will know far more about this than me!

 

Annie

Share this post


Link to post
Share on other sites

Just since this was mentioned a couple of times in this thread... There is no such thing as selective color in GIMP. That's why I stopped using GIMP (at least for the NB processing) long ago. I was landing in this thread because I was actually hoping that somebody has sufficient experience with GIMP 2.10 in the meanwhile to explain how to get the typical Hubble gold and turquois from SHO data.

cs, Pat

Share this post


Link to post
Share on other sites
Guest

+1 for keeping the green in there somewhere - really like your second of two examples ?

David

Share this post


Link to post
Share on other sites

I have never liked green in astro images and hence don't like SHO.... I think that by the way people are trying to eliminate it from your image I am not the only one lol.

However I agree that SHO should be green and so changing it really misses the point in assigning the channels like that.....sure you can then manipulate it into something else that is in itself valid but it is not SHO any more. 

If someone presents an image with the pallet specified as SHO then I think everyone needs to expect it to be green and evaluate it as a true SHO image. If you manipulate it to the point when the Ha is orange then its not SHO irrespective of how you originally assigned the channels.

As a result of this I will say that I think the original is a great example of a SHO image. 

Adam 

Edited by Adam J

Share this post


Link to post
Share on other sites

Ok, this was long overdue, I promised raw files for all interested to have a go.

I was trying to find time to implement all the changes in data reduction algorithms, and it's not finished yet, so here are current raws (no sigma clip with weighted average yet, Ha contains some very faint satellite trails)

H alpha:

http://serve.trimacka.net/astro/2018-10-13/stack0000.fits

SII:

http://serve.trimacka.net/astro/2018-10-13/stack0001.fits

OIII:

http://serve.trimacka.net/astro/2018-10-13/stack0002.fits

And here is my latest effort - can't be called modified SHO, it is exactly SHO (no channel mixing):

bubble_again.png

With few "tricks"

- all data stretched prior to mixing

- SHO -> RGB mapping

- G reduced to below 20% in intensity (other components remained the same and no inter channel mixing)

- Magenta desaturated 100%

- Lab color transfer to Ha as Lum

(seasoned to taste with sharpening and other gismos). SNR is very poor so it all looks "dull" but if stretched more - too much noise appears.

  • Like 1

Share this post


Link to post
Share on other sites

Here's my attempt with your data. I just downloaded v 2.10 of GIMP and I used your data set to get my feet wet with GIMP processing.

Boy, is this a learning curve! I think I will stick with PixInsight. Much easier. ?

vlaivbubble_rgb-compose.jpg.cbd9ab40a5d83ca083af8595d76f937f.jpg

Basically:

- load 3 masters as layers

- levels for first stretch

- curves for second stretch

- bringing in the blackpoint and tweaking with levels and curves alternated

- aligning the histograms

- channel combination

- saturation and lightness adjustment of green and magenta

- duplicated layer and created a lightness mask for the top layer

- increased contrast in the lightness mask

- noise reduction on the bottom layer

- blended the two layers (soft light), rescaled, and saved as jpeg

Share this post


Link to post
Share on other sites

And this mine PI interpretation ;)

Made quite a lot of adjustments: main steps, deconv, stretched each manually up to the same level, SHO combination (all 100%), used Color Mask script to create masks per channel (RGBY), extracted L and curved with slight S. Played with Masks and curves to get colors I want, - applied LRGB combination with Luminance to shapen more and once again with Chrominance Noise Reduction.

Most difficult part... - Created mask for Most bright stars and the Spyke one, - reduced halos.

TGV denoise with following DarkStructure enhance script.

Anyway... I always "kill" star sharpness :( still learning ;)

Bubble_V2.thumb.jpg.430b6fe6cf07e7bf62fa871fa5242891.jpg

 

Share this post


Link to post
Share on other sites
2 minutes ago, RolandKol said:

And this mine PI interpretation ;)

Made quite a lot of adjustments: main steps, deconv, stretched each manually up to the same level, SHO combination (all 100%), used Color Mask script to create masks per channel (RGBY), extracted L and curved with slight S. Played with Masks and curves to get colors I want, - applied LRGB combination with Luminance to shapen more and once again with Chrominance Noise Reduction.

Most difficult part... - Created mask for Most bright stars and the Spyke one, - reduced halos.

TGV denoise with following DarkStructure enhance script.

Anyway... I always "kill" star sharpness :( still learning ;)

Bubble_V2.thumb.jpg.430b6fe6cf07e7bf62fa871fa5242891.jpg

 

Like the colors!

Try using H alpha as luminance - it's got best SNR of the bunch and also looks the sharpest (smallest stars) - that way you will not have to do much sharpening and faint background will contain less noise.

Share this post


Link to post
Share on other sites

And back in my comfort zone, here's the data processed with PixInsight

modified SHO with Ha contributing to luminance

vlaivBubble_PI.thumb.jpg.924cf578c6ab7e2534cca0a187b6a5ad.jpg

Share this post


Link to post
Share on other sites

My own take on this is that the 'official' Hubble palette shows as predominantly green and this can be seen in many famous HST images, the gold and turquoise renditions that most amateur astronomers produce (including me!) are an artistic impression of the true data. I really like Wim's PI version. For the fun of it, here's my go at your data using PhotoShop.:-

Official Hubble Palette

1651983074_HubbleGreen.png.4777f204c497d67abadfb5ab75870a63.png

Turquoise and Gold version

Hubble_Gold.png.0cc4238b9c40db233690e7a2f46388ad.png

  • Like 1

Share this post


Link to post
Share on other sites

I think that we need to dig deeper into Hubble palette as used by Hubble team.

As far as I can see, it has changed over the years, as this comparison of Pillars clearly shows:

image.png.3c417031bdde3c9212e382de0175c2ab.png

Old style rendition indeed shows distinctive green cast, while newer has much more subtle tones (wonder if Hubble team is following the trends set by peoples expectations for SHO - removal of green cast).

Luckily we can go directly to the source and inspect scientific grade fits used to compose images like above - in this particular case, since this is thread on Bubble nebula, I've found download section for this particular object (alongside reference rendition, which does not show pronounced green cast):

https://archive.stsci.edu/prepds/heritage/bubble/introduction.html

There are two kind of fits available - sci and weighted, second one, I presume, used to compose image on the page (Hubble Bubble rendition).

I'm thinking about downloading both version and doing quick comparison to try to establish in what ways were they altered and if they compose to give distinct cast (any stretch and enhancements needed).

 

Share this post


Link to post
Share on other sites

Just a quick update, after 3GB download from above link - weight fits are not what I thought they were :D

But it turns out that basis for their image composition is NHO (they captured NII data not SII) is 1:0.2:1 ratio also without channel mixing, so I was pretty close with my above color rendering (I also lowered Ha to somewhere around 20% or a bit lower).

Share this post


Link to post
Share on other sites

Thats IT!
I am buying monitor for my Notebook!!!!

I just saw my image from the office PC... Sooooo UGLY !!!! :) And it looked sooo beautiful on the Notebook! :)
Insane! :)

 

 

  • Haha 1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.