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.

sgl_2019_sp_banner.thumb.jpg.a0ff260c05b90dead5c594e9b4ee9fd0.jpg

andyo

flaming star nebula(IC405)

Recommended Posts

Hi 

Started this one back in December but due to work commitments and lack of clear skies only finished it last week. I did both a narrowband and HaRGB. I found there was hardly any O111 for this target. Maybe the narrowband is a little to soft with the processing not to sure myself.

Narrowband Ha,O111 and S11 all 12x900secs

RGB 15x 300secs

5a9072d2e62e1_Flamingstarnebulanarrowband(IC405)December2017.thumb.jpg.48a1420f6c9cf40d8c6980efcc366783.jpg

5a9072ea31403_flamingstarnebula(ic405)December2017.thumb.jpg.b57601510e1a76484e9fa5c535c8d002.jpg

 

  • Like 13
  • Thanks 1

Share this post


Link to post
Share on other sites
30 minutes ago, andyo said:

Started this one back in December but due to work commitments and lack of clear skies

I know exactly how that feels... too much cloud and life wasting work.. not enough astronomy.... IMHO

  • Like 1

Share this post


Link to post
Share on other sites
24 minutes ago, MarsG76 said:

Very nice result... what a great looking object, well done.

 

Thanks MarsG76

19 minutes ago, MarsG76 said:

I know exactly how that feels... too much cloud and life wasting work.. not enough astronomy.... IMHO

Ha Yes and then you get the clear night but are just to knackered after a life wasting day in work

13 minutes ago, des anderson said:

Beautiful detail,you must be proud of these images. Thanks for sharing. Des

Thanks a lot Des, yes I am pleased with the way these turned out especially the narrowband one 

  • Like 1

Share this post


Link to post
Share on other sites

Very well done Andy - very pleasing results.  I especially like the HaRGB version and you have captured the blue reflection nebula very well indeed.

For constructive feedback, there is some detail within the nebula that is not present or crisp as you note.  It may be from focus or perhaps processing steps.  However, please don't be demotivated by this.  Recognising areas for improvement is a key step in becoming a reflective practitioner.

If you want to pursue a discussion on processing steps, post into your thread and you get plenty of tips I'm sure :icon_biggrin:.

Edited by Barry-Wilson
typo
  • Like 1

Share this post


Link to post
Share on other sites
36 minutes ago, Barry-Wilson said:

Very well done Andy - very pleasing results.  I especially like the HaRGB version and you have captured the blue reflection nebula very well indeed.

For constructive feedback, there is some detail within the nebula that is not present or crisp as you note.  It may be from focus or perhaps processing steps.  However, please don't be demotivated by this.  Recognising areas for improvement is a key step in becoming a reflective practitioner.

If you wabt to pursue a discussion on processing steps, post into your thread and you get plenty of tips I'm sure :icon_biggrin:.

Thanks Barry,I welcome all constructive feedback It helps us improve our images. I believe the detail was lost in the processing maybe by overdoing noise reduction or possibly with gradient removal. Processing is something I am continually working on (recently purchased Steve great book dark art or magic bullet).I am using Pixinsight for callibration DBE and an initial stretch then into photoshop for further processing such as more levels, curves sharpening satuaration and noise reduction. However on this target I still wasn't happy with the background gradient despite using DBE earlier in the process, so applied gradient xterminator maybe sacrifising some detail in the process.

Share this post


Link to post
Share on other sites
On 24/02/2018 at 12:38, Barry-Wilson said:

Very well done Andy - very pleasing results.  I especially like the HaRGB version and you have captured the blue reflection nebula very well indeed.

For constructive feedback, there is some detail within the nebula that is not present or crisp as you note.  It may be from focus or perhaps processing steps.  However, please don't be demotivated by this.  Recognising areas for improvement is a key step in becoming a reflective practitioner.

If you want to pursue a discussion on processing steps, post into your thread and you get plenty of tips I'm sure :icon_biggrin:.

Hi Barry I reprocessed this and would be interested if you think this is an improvement or not

5a92b5fde9b0f_Flamingstarnebulanarrowband(IC405)December2017.thumb.jpg.112e75609c4e38ce2c8fb6441cdfbb31.jpg

 

  • Like 1

Share this post


Link to post
Share on other sites

Hi Andy.

My eye is drawn firstly to the dark ringing around your stars.  This artefact can arise after deconvolution and/or star shrinking (or replacement) and it seems more pronounced in this latest version than in your original.

Sometimes the data responds better to 'less' processing - my experience is solely in PixInsight and I'll happily help with the steps that you have outlined above.  If you'd also like me to help with more detailed non-linear processing in PI just ask too, it's no trouble :happy11:.

Some initial questions (screenshots of your steps will help me understand):

1. How are you stacking your individual channels?  Are you using Image Integration or relying on 'Masters' from BPP or are you calibrating each step?

2. How are you combining your 3 NB channels?

3. How are you applying DBE, eg what settings, how many data points?

4. How are you carrying out your initial stretch on the combined linear SHO?

5. Are you applying any SCNR?

6. Are you using the Ha channel as a synthetic luminance?

  • Like 1

Share this post


Link to post
Share on other sites

Hi Barry

Thanks for the response much appreciated. I have outlined my processing procedure below

After using the Batch processing script to calibrate the individual subs I then use the image integration module to stack the individual frames. After a crop on the three masters i use DBE. I am manualy placing the points in DBE due to the amount of nebulosity using 12 as the size and subtraction as the correction method.weight down to 650 and tolerance up to 7. 

The initial stretch in this instance was applied by transferring the settings from the screen transfer function to the histogram transformation module. I used statistics to check that the mean background was more or less the same at 0.25. I saved each individual master as a 16bit tiff and then opened them in photoshop. Here I combined the three masters into a SHO using the merge channels channels RGB setting. I then used the new adjustment layer to create 4 new selective colours and used the settings outlined in Steves book to adjust the colour.I ran this twice. Then I created a star layer and reduced the star size using the minimum filter keeping the radius at 1 after which I faded this to 50%.

I then adjusted the levels by moving the black point using several iterations keeping each one to about 5 and being careful not to clip it. I also used curves to add contast and then using a starlayer sharpened it using smart sharpen then applied an inverted mask and used the paintbrush to paint in only those bits I wanted sharpening. I then used match colour to increase the colour intensity slightly. I then used the photoshop plugin noiseware to decrease the noise slightly then adjusted the blackpoint in levels. Finally I transferred the image back into PI for SCNR setting on average neutral and amount set to about 50.

Ha stretched.tif

O111 stretched.tif

S11 stretched.tif

Above are the three stretched masters saved as a TIF 

 

Share this post


Link to post
Share on other sites

OK - to take this a PI processing step at a time.

1. At Image Integration stage, I have been using the new Local Normalisation process and have found this step helpful in flattening the background.  This is an optional step though, we can discuss later.

2. For myself, after a Dynamic Crop of the Ha, SII and OIII stacks, I would firstly use PixelMath to combine the stacks into the SHO master.  The creation of the SHO master is best carried out at the linear stage.  Do you know how to do this?

3. After creating the SHO master I would then use the STF tool, with channels unlinked.

4. Then use DBE on the SHO master and Ha master, the same instance applied.  I would leave the DBE settings at default for the time being, place the samples manually as you suggest (close as possible to corners, samples on the edges and across the frame), subtraction as correction method.

5. The STF to Histogram Transformation transfer method will work well for the SHO master, however for the Ha master I would use the HT tool with multiple iterations.  The non-linear Ha will be very useful to provide a 'luminance' for the polishing of the final SHO image.

6. After carrying out the STF to HT stretch of the SHO I would then use SCNR set to Green.  Experiment with the amount to remove; try removing 80%.

7. After stretching the Ha master, try adding this to the SHO master using Channel Combination set to CieLab - untick 'a' and 'b' and select the non-linear Ha as luminance (you may prefer to execute this step in PS or whatever the equivalent process may be in PS).

8. Then save the files as the format you need for PS.

9. Optional steps - are the use of Mure Denoise as a first step, Local Normalisation prior to Image Integration, Deconvolution on the Ha master, HDR Multiscale Transform and Local Histogram Equalisation on the Ha master, Colour Masks for adjustments to the colour palette along with Curves etc - ooh loads of optional steps and processes :icon_biggrin:.  I'm happy to help with these further more advanced steps if you wish.

HTH.

  • Like 1

Share this post


Link to post
Share on other sites

Thanks a lot for this help Barry.

The new local Normalisation is something I have not come across before. Would be interesting to look into this.

No I don't know how to create the SHO master at the linear stage using pixel math I usually do this in PS after the stretch

Can I ask why you unlink the channels before using DBE as I thought the STF tool had no actual effect upon the image itself and was purely so you could see the image to work on it.

Colour masks for adjustments to the colour palette sounds interesting also. 

I have used local histogram equalisation before but am unsure about this i never seem to be able to get it right 

Share this post


Link to post
Share on other sites

I forgot to ask Barry how would you carry out star reduction

Share this post


Link to post
Share on other sites

Pixel Math to create SHO linear:

1. With PixelMath open, untick the Use a single RGB/K expression' option.  This allows you to assign different stacks to each RGB channel.

2. Tick Create New Image. Name your image in the empty Image ID box.

3. Select Colour in the 'Colour Space' box.

4. Apply.

When you use the STF it doesn't make a permanent change to the linear files, it is only a display.  However, you are tranfering the settings used in the STF to the HT to make a permanent stretch.  The settings in the STF will be different depending on whether the channela re linked or unllinked. Try it with both and see.

It's best to master the basic workflow steps before we move on.

Have you watched Harrys' tutorials?  If not, I thoroughly recommend that you do.

 

  • Like 1

Share this post


Link to post
Share on other sites
3 minutes ago, Barry-Wilson said:

Pixel Math to create SHO linear:

1. With PixelMath open, untick the Use a single RGB/K expression' option.  This allows you to assign different stacks to each RGB channel.

2. Tick Create New Image. Name your image in the empty Image ID box.

3. Select Colour in the 'Colour Space' box.

4. Apply.

When you use the STF it doesn't make a permanent change to the linear files, it is only a display.  However, you are tranfering the settings used in the STF to the HT to make a permanent stretch.  The settings in the STF will be different depending on whether the channela re linked or unllinked. Try it with both and see.

It's best to master the basic workflow steps before we move on.

Have you watched Harrys' tutorials?  If not, I thoroughly recommend that you do.

 

Hi Barry

Yes I have watched Harrys' tutorials although it was a while back now 

i am just having a go of the local normalisation tool and then going to work through the tips you have given to see how I get on with them.

Share this post


Link to post
Share on other sites

Hi Barry I have followed the above workflow you gave (including local normalisation) I also applied some local histogram equalisation to the master Ha. At point 6 above I moved both the master Ha and master SHO into PS and combined the master Ha as a luminance onto the master SHO at 50% opacity.

I used the selective colour process Steve suggests in his book dark art or magic bullet. 

Applied some curve adjustments and a smart sharpen.

Then back into PI for morphological transformation to reduce the star sizes before finally saving. 

I think this is a big improvement a lot more pleasing on the eye as well as a more natural appearance to it and no dark halos around the stars, what do you think?

Thanks for all the help you have given me with this it is very much appreciated

5a932097a709f_Flamingstarnebulanarrowband(IC405)December2017.thumb.jpg.9a21d1b6135e3b9915fb5756c20db9cb.jpg

 

  • Like 1

Share this post


Link to post
Share on other sites

This latest image is much improved: well done.

Structure and detail has been preserved and there is a more harmonious SHO palette.

You now need to work on the background and star colour (removing magenta) and some gentle star reduction.  I would also encourage you to carry out the Ha lum addition as CieLab luminance in PixInsight as it will help resolve your background sky.

I’m out tomorrow evening but will look in Tuesday to see how you are getting on.

You clearly have good data and I hope you are delighted with the developments in your processing techniques.

Barry

  • Like 2

Share this post


Link to post
Share on other sites
2 minutes ago, Barry-Wilson said:

This latest image is much improved: well done.

Structure and detail has been preserved and there is a more harmonious SHO palette.

You now need to work on the background and star colour (removing magenta) and some gentle star reduction.  I would also encourage you to carry out the Ha lum addition as CieLab luminance in PixInsight as it will help resolve your background sky.

I’m out tomorrow evening but will look in Tuesday to see how you are getting on.

You clearly have good data and I hope you are delighted with the developments in your processing techniques.

Barry

Hi Barry

I certainly am pleased with the improvements. I will have a go at the pi lum addition. What do you suggest for star reduction I used morphological transform to reduce the size. As for reducing the magenta star colour and background how would you do this in pi. As ever appreciate your help

Share this post


Link to post
Share on other sites

Hi Andy

Last post before bed :happy10:.

There is a downloadable process for removing magenta stars (it’s a Pixel Math formula). Do a Google search and then download it. You can the use it by using the Favourites menu.

Star reduction will take some time to explain & I don’t have time now. A star mask which encompasses stars of all sizes is key.

Then we can look at special masks to assist colour boosting and LHE.

TTFN

Edited by Barry-Wilson
  • Like 1

Share this post


Link to post
Share on other sites

The redone NB version is now very classy. My heart lies in broadband imaging so I do prefer the HaRGB, especially on this target where there is the interesting reflection nebula to be had in the blue channel. You've separated this nicely from the Ha signal, something which isn't always easy.

Olly

Share this post


Link to post
Share on other sites
11 hours ago, ollypenrice said:

The redone NB version is now very classy. My heart lies in broadband imaging so I do prefer the HaRGB, especially on this target where there is the interesting reflection nebula to be had in the blue channel. You've separated this nicely from the Ha signal, something which isn't always easy.

Olly

Thanks Olly. i was surprised how well the reflection nebula showed as there seemed to be so little of it on the subs

Share this post


Link to post
Share on other sites
21 hours ago, Barry-Wilson said:

Hi Andy

Last post before bed :happy10:.

There is a downloadable process for removing magenta stars (it’s a Pixel Math formula). Do a Google search and then download it. You can the use it by using the Favourites menu.

Star reduction will take some time to explain & I don’t have time now. A star mask which encompasses stars of all sizes is key.

Then we can look at special masks to assist colour boosting and LHE.

TTFN

Hi Barry had a search for the magenta star reduction using pixel math and there was a post on the pixinsight forum where someone using the user name troypiggo had attached a zip file with a formula but it appears the zip file is no longer there.

However I did find this formula

 R/K: $T
    G:  m = min($T[0], $T[2]); iif(m>$T[1], m, $T)
    B: $T

symbols: m

so made a process icon which I have saved and it seems to be very effective. The result is as below

5a946eb149a46_flamingstarwithmagentastarreduction.thumb.jpg.932218cf231ae29af41562eb9007e0cd.jpg

This is where I am at now, so will wait for your input and take it from there, thanks again Barry your help has been immense

Share this post


Link to post
Share on other sites

Hi Andy

Looking very good & certainly much improved.

To help the background colour there are a couple of simply options:

1. Carry out a further DBE using a few samples, eg 6 or 7 across the background sky.

2. Extract the luminance and using that as a mask, protect the nebula and stars, and the reduce the saturation of the unprotected background sky using Curves.  You may need to invert the mask afterwards to boost the saturation of the nebula.

Barry

  • Like 2

Share this post


Link to post
Share on other sites
1 minute ago, Barry-Wilson said:

Hi Andy

Looking very good & certainly much improved.

To help the background colour there are a couple of simply options:

1. Carry out a further DBE using a few samples, eg 6 or 7 across the background sky.

2. Extract the luminance and using that as a mask, protect the nebula and stars, and the reduce the saturation of the unprotected background sky using Curves.  You may need to invert the mask afterwards to boost the saturation of the nebula.

Barry

OK Barry will try that tomorrow. I didn't realise you could use DBE in the non-linear stage. Out braving the weather walking the dog now 

Share this post


Link to post
Share on other sites

Don’t you two think you are having your own conversation here as I am taking notes every step of the way ?, and I’m sure I am not alone!

Thank you both for a very informative read so far and especially to Barry for his hints and tweaks, some of which never even occurred to me, such as reducing background sky saturation with curves - I have only ever darkened the sky slightly with rgb/k so definitely one for me to try. 

If I may add one thing re magenta stars; my normal first try method now is to invert the image and use scnr at around 0.9. Inverting the image turns the magenta stars green which is where scnr kicks in. Simply then reinvert the image. This isn’t as sophisticated as the formula (which I do use as well) but usually gives a pleasing result. 

Thanks to both again. 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • 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.