Too many stars - what and how to cope with it? - Forums

Taking deep images on a region containing SH2-147 - LRGB plus Ha,OIII and SII results in a problem - in that area, the stars are very dense. Extracting stars, processing the nebulous area separately is clear but after that - what to do? There are so many stars, too many stars. I tried to alter the stretching to stretch only the brightest stars, OK, it seems to work but it does not make me happy what I can see. Stretching the stars more or less in a normal way - no way, the stars cover more or less completely the nebulous area.

I know, I need to take more OIII and SII subs to get the nebulous area better than now, but I think one can see already now, what I am aiming at.

What would you do in such a case? How would you solve that problem? I have attached two images - one with the reduced amount of stars and one with the normal amount of stars (which is a no-go IMO).

📷 SH2-147_QHY268M_CFW_HG6_normalstars1201125.jpg📷 SH2-147_QHY268M_CFW_HG6_reducedstars121125.jpgThanks for your advise!

CS
Georg

What's the green? Beside that, it's full of holes and very noisy. What was it taken with, scope wise?

And, yes, you can make the stars far less intrusive but avoiding the issues with the second image won't be easy.

For star “reduction” i simply stretch the stars a bit less. If their glow is a bit too bright still, I pull the RGB curve down in the center slightly. If that’s STILL too much, a mild touch of Morphological Transformation can be applied to the star layer.

andrea tasselli · Nov 12, 2025, 05:46 PM
What's the green? Beside that, it's full of holes and very noisy. What was it taken with, scope wise?

That is correct - it is noisy because it is not deep enough and stretched a lot. The green - that is SII (HSO). The scope - well it is the Sharpstar 150, f/2.8 at 420mm FL with a QHY268M and LRGB HSO filters.
I know, I need much more time to make it good but what bothers me is the huge amount of stars (they are LRGB)

When I process images with dense star fields, I use BlurX to make the stars a little smaller, then StarX to extract the stars, and then MultiscaleMedianTransform to remove the smallest stars (1x and 2x, usually). Then I stretch as desired and recombine with the starless image.

Georg N. Nyman:
That is correct - it is noisy because it is not deep enough and stretched a lot. The green - that is SII (HSO). The scope - well it is the Sharpstar 150, f/2.8 at 420mm FL with a QHY268M and LRGB HSO filters.
I know, I need much more time to make it good but what bothers me is the huge amount of stars (they are LRGB)

*Not sure what you're aiming at (am kind of liking this approach to SHO though) but the best way to proceed is to use RGB stars (NOT LRGB) only to start with and add them in the starless image once stretched with Asinh and operating on the intensity of the star field using the HistogramTransformation process. You'd iterate between the two to find the right intensity that covers the holes but still leaves the star field unobtrusive.

andrea tasselli · Nov 12, 2025, 07:35 PM
Georg N. Nyman:
That is correct - it is noisy because it is not deep enough and stretched a lot. The green - that is SII (HSO). The scope - well it is the Sharpstar 150, f/2.8 at 420mm FL with a QHY268M and LRGB HSO filters.
I know, I need much more time to make it good but what bothers me is the huge amount of stars (they are LRGB)

*Not sure what you're aiming at (am kind of liking this approach to SHO though) but the best way to proceed is to use RGB stars (NOT LRGB) only to start with and add them in the starless image once stretched with Asinh and operating on the intensity of the star field using the HistogramTransformation process. You'd iterate between the two to find the right intensity that covers the holes but still leaves the star field unobtrusive.

Good idea - but why only RGB stars and not LRGB for the stars?

Georg N. Nyman:
andrea tasselli · Nov 12, 2025, 07:35 PM
Georg N. Nyman:
That is correct - it is noisy because it is not deep enough and stretched a lot. The green - that is SII (HSO). The scope - well it is the Sharpstar 150, f/2.8 at 420mm FL with a QHY268M and LRGB HSO filters.
I know, I need much more time to make it good but what bothers me is the huge amount of stars (they are LRGB)

*Not sure what you're aiming at (am kind of liking this approach to SHO though) but the best way to proceed is to use RGB stars (NOT LRGB) only to start with and add them in the starless image once stretched with Asinh and operating on the intensity of the star field using the HistogramTransformation process. You'd iterate between the two to find the right intensity that covers the holes but still leaves the star field unobtrusive.

Good idea - but why only RGB stars and not LRGB for the stars?

Because, beside being more colourful, the are also less prominent by and large so easier to tame in compositions.

andrea tasselli · Nov 12, 2025, 07:47 PM
Because, beside being more colourful

This sounds like a processing error, properly done LRGB should not change the colour intensity.

From what I can read here and by looking at other images of the same area, I have done something very wrong… can happen, I guess!
I need to review the entire collection of subs and maybe have to start from scratch again…

Thanks for your feedback and suggestions, they helped a lot!

An effective way to deemphasize a dense star field is to use MorphologicalTransform to “erode” the size and brightness of the field. Using BXT first makes the stars sharper but you have to be careful that the MT erosion function doesn’t complete remove the really faint stars. You can use the strength setting to control how strongly the “Erode” function affects the star field. MT erosion provides a lot of control over the relative brightness of a star field so try it out.

John

I am hesitating to post this now, that´s how much I feel ashamed for what I just found out by checking the equipment - I used the wrong filter wheel. No idea how this could happen, but it did.

I was wondering all the previous weeks, why I do not get more OIII and Ha photons despite f/2.8 and the QHY268M… well no wonder, I used the 50mm filter wheel with the standard filter, not that for f/2!

Well, that explains also a lot - as Andrea asked, why is the image so noisy ( I had to stretch it far too much).. that made me think about why really? And so I found out, that I used all the time the wrong filters.

Oh well, I think I learned a lesson - but regarding the stars, thanks a lot, I tried already the Morphological Transformation and yes, it works very fine on the stars.
OK, back to the drawing board, waiting for clear skies and restart!

Thanks to all, CS
Georg😩

What are the exposure times of your RGB? Shortening the time of RGB subs can reduce saturated stars and let you control the star size more easily.

Bill blanchard has an excellent set of star reduction scripts for pixinsight.

The way I overcome dense star fields is, I shoot separate stars of much shorted exposures, even for broadband targets.

This allows me to get a lower number of stars for focusing on the target more as well as have the full star field for mouse over. I like this because it gives me the best of both worlds.

Example using this process (open the link and hover over image for full stars version):

📷 The Shark Nebula - LDN 1235 (Reduced Stars)

https://app.astrobin.com/i/v668my/

Georg N. Nyman · Nov 12, 2025, 10:24 PM
I am hesitating to post this now, that´s how much I feel ashamed for what I just found out by checking the equipment - I used the wrong filter wheel. No idea how this could happen, but it did.
I was wondering all the previous weeks, why I do not get more OIII and Ha photons despite f/2.8 and the QHY268M… well no wonder, I used the 50mm filter wheel with the standard filter, not that for f/2!
Well, that explains also a lot - as Andrea asked, why is the image so noisy ( I had to stretch it far too much).. that made me think about why really? And so I found out, that I used all the time the wrong filters.
Oh well, I think I learned a lesson - but regarding the stars, thanks a lot, I tried already the Morphological Transformation and yes, it works very fine on the stars.
OK, back to the drawing board, waiting for clear skies and restart!
Thanks to all, CS
Georg😩

Sorry to hear that George but don’t feel ashamed! That is just one of the 2,647 ways that I’ve pioneered screwing things up over the years. And, I’m not done. I’m still coming up with new and even more innovative ways to screw things up! So, be proud. You are just following in the wake of those of us who have gone before, leaving piles of wasted subs scattered all along the side of the road. :))))

Good luck getting back on track!

John

Star reduction if you are using PixInsight. Works wonders!

Need I mention star reduction method from Bill Blanshan? Basically his script does a destretch of the stars as some here are stating. However, the advantage of using the script is that one can so easily try a bunch of different stretches of the stars on the final image in seconds and rapidly compare the results. That way you can pick the most desirable result, the one that suits you. Requires that your image can sustain a good result when using the star removal method of choice. The script uses the final destarred image, and then the starfield image extracted by SXT, or the like to recombine the stars to the final image at any stretch very easily. Heck, if you even want to make the stars brighter than you have before extraction, you can even do that! Anyway, the whole process becomes a task that takes but a few minutes of time, rather than the pain of trying to stretch your stars perfectly in the stars only image, then have to recombine, then judge, etc. Yes, I am sure many here are wonders with pointing clicking, etc., but all that is unnecessary. Also, the stars come from the image that they are intended to repopulate, so you do not tend to see any issue with halos (bright or dark) around the final image, such as I have seen regularly here from stars that were extracted early in the processing only to be combined after many iterations of processing the starless field.

Alan Brunelle · Nov 13, 2025, 04:28 AM
Need I mention star reduction method from Bill Blanshan? Basically his script does a destretch of the stars as some here are stating. However, the advantage of using the script is that one can so easily try a bunch of different stretches of the stars on the final image in seconds and rapidly compare the results. That way you can pick the most desirable result, the one that suits you. Requires that your image can sustain a good result when using the star removal method of choice. The script uses the final destarred image, and then the starfield image extracted by SXT, or the like to recombine the stars to the final image at any stretch very easily. Heck, if you even want to make the stars brighter than you have before extraction, you can even do that! Anyway, the whole process becomes a task that takes but a few minutes of time, rather than the pain of trying to stretch your stars perfectly in the stars only image, then have to recombine, then judge, etc. Yes, I am sure many here are wonders with pointing clicking, etc., but all that is unnecessary. Also, the stars come from the image that they are intended to repopulate, so you do not tend to see any issue with halos (bright or dark) around the final image, such as I have seen regularly here from stars that were extracted early in the processing only to be combined after many iterations of processing the starless field.

Thanks a lot Alan, I was not aware that such a script exists… I downloaded it and it is great!

John Hayes · Nov 13, 2025, 03:32 AM
Georg N. Nyman · Nov 12, 2025, 10:24 PM
I am hesitating to post this now, that´s how much I feel ashamed for what I just found out by checking the equipment - I used the wrong filter wheel. No idea how this could happen, but it did.
I was wondering all the previous weeks, why I do not get more OIII and Ha photons despite f/2.8 and the QHY268M… well no wonder, I used the 50mm filter wheel with the standard filter, not that for f/2!
Well, that explains also a lot - as Andrea asked, why is the image so noisy ( I had to stretch it far too much).. that made me think about why really? And so I found out, that I used all the time the wrong filters.
Oh well, I think I learned a lesson - but regarding the stars, thanks a lot, I tried already the Morphological Transformation and yes, it works very fine on the stars.
OK, back to the drawing board, waiting for clear skies and restart!
Thanks to all, CS
Georg😩
Sorry to hear that George but don’t feel ashamed! That is just one of the 2,647 ways that I’ve pioneered screwing things up over the years. And, I’m not done. I’m still coming up with new and even more innovative ways to screw things up! So, be proud. You are just following in the wake of those of us who have gone before, leaving piles of wasted subs scattered all along the side of the road. :))))
Good luck getting back on track!
John

i wasted a rare clear and relatively moonless night this week. My lumin and blue filter run were set up perfectly ,somehow i set my green and red runs were set to 5 seconds ..

I don’t know how it happened :)

in pixinsight: the star reduction script by @Bill Blanshan and @Mike Cranfield - https://cosmicphotons.com

several approaches to play with …

The SetiAstro Star Stretch script, aside from from doing a pretty good job at stretching stars, also has the ability to reduce or inflate them, and lift or lower saturation. It doesn’t have the range of Options that the Blanshan and Cranfield script does, but it works well and is easy to use if you don’t need the bells and whistles.

Forget the RGB stars. Take your Ha and OIII stars and use Frank’s (SetiAstro) NarrowbandToRGBstars script to create decent looking stars. True, they are not going to be absolutely SPCC color accurate, but they look pretty darn nice. If I’m shooting a heavy nebulosity target, that is what I do these days. Look at these HOO stars in this image.

https://app.astrobin.com/u/AlnitakProductions?i=zlr2xh#gallery

If it makes you feel any better, I've been hit on the head by my camera.

Twice