Back to forums

Iris Nebula

Request for feedback NGC7023 Deep Sky Nebula NGC Objects
17 replies•811 views
Dimitris Kavallieratos avatar
Dimitris Kavallieratos avatar
Hello !

Throughout galaxy season I have invest a lot of time-due to focal length shortage so there are not good targets for me this time of the year- on the Iris Nebula and with ~12 hours of good data from Bortle 8 ,I have a kind of acceptable result.

Should I invest more time on this target and have a >20 hrs integration time or it is just a loss of time from there on? That dark nebula that surrounds the Iris is barely visible on my integration, how is it possible that people with about the same integration time as me and LP have achieved fantastic results? I shoot with my ASI533 OSC and L-Pro filter with exposures of 180s and 300s.

Thank you
Christoph Nieswand avatar
Christoph Nieswand avatar
Hi Dimitris
For Bortle 8 this is a good result, I think. Did you have moonlight in your subs?
If you compare to others … check their Bortle scale …
I mean Bortle 8 is really a serious thread!!!
If it is worth going for more integration time? … Well it would certainly help … but this is a lot of time … depends on you, if you want to invest it.

Chris
Dimitris Kavallieratos avatar
Dimitris Kavallieratos avatar
Christoph Nieswand:
Hi Dimitris
For Bortle 8 this is a good result, I think. Did you have moonlight in your subs?
If you compare to others ... check their Bortle scale ...
I mean Bortle 8 is really a serious thread!!!
If it is worth going for more integration time? ... Well it would certainly help ... but this is a lot of time ... depends on you, if you want to invest it.

Chris

Hi Chris!

Unfortunately yes , I had moonlight in a night or two…

I am waiting for some targets suitable for my fov to rise high in the sky, so there is not something great to shoot in the coming weeks( I think)! 

If getting more time will improve my final image and bring the dark nebula out ,then I will do it.It is and awful lot of time indeed…
Mike Cranfield avatar
Mike Cranfield avatar
Nice work Dimitris - this is a great looking image already!

My thoughts, FWIW,  are that there may be more you could get from the data without needing more integration time.  It looks to me like there may be some remaining gradients - you have applied DBE which is a great too but can be quite challenging to refine and find the best parameters - in my experience it repays some fiddling to achieve the best result.  There is a fair amount of detail you have captured in the surrounding nebulosity - you have tried using curves to introduce contrast, have you tried using the Generalised Hyperbolic Stretch script?  This allows you to do some very targeted stretching to bring contrast into specific areas of your image.  Dave Payne and I are presenting GHS on the Astro Imaging Channel tomorrow which will give some background to how best to use it.  I have not used NoiseExterminator but there are some great reports of it and it certainly seems to have done a nice job on your image here.  It may just be my eyes, but it looks like there may be a slight green cast to your image, a dose of SCNR might help clean that up.

Having said all that, there are many far more experienced image processors than me on this forum who I am sure can give you some much more expert pointers to maximise the potential of the data you have.

Good luck and CS

Mike
Helpful Respectful Engaging Supportive
andrea tasselli avatar
andrea tasselli avatar
Hello !

Throughout galaxy season I have invest a lot of time-due to focal length shortage so there are not good targets for me this time of the year- on the Iris Nebula and with ~12 hours of good data from Bortle 8 ,I have a kind of acceptable result.

Should I invest more time on this target and have a >20 hrs integration time or it is just a loss of time from there on? That dark nebula that surrounds the Iris is barely visible on my integration, how is it possible that people with about the same integration time as me and LP have achieved fantastic results? I shoot with my ASI533 OSC and L-Pro filter with exposures of 180s and 300s.

Thank you

I think it is a hell of the job with the sort of skies you have and you should be proud of yourself.  Adding data is always worth it as it allows to ease on the NR pedal. This said you would want to remove the copyright right in front of the nebula so we can see it! Stars colours need improving too. Otherwise pretty good going.
Taman avatar
Taman avatar
Hi Dimitris,

For a Bortle 8 sky this is a good result, but there is far more dust visible in your image than you think! If you look at the histogram, it is too far over to the right and quite narrow, also the image looks a bit washed out. Just for a bit of fun I converted your image to 16 bit and ran it through Starnet++ to remove the stars. I then gave it a quick stretch in Siril, before setting the black point to 25,25,25 in Photoshop levels. I used a point halfway between the nebula and the bottom right corner which worked well. Finally I reduced the stars slightly before adding them back in. It's not perfect, but it gives an idea of what is hiding in there! Sorry for the poor quality but I only had your 8 bit jpeg to work with. You should get far better results with your original data.

Hope this helps.

Cs, Tony.

iris.jpg
Well Written Helpful Insightful Respectful Engaging Supportive
Dimitris Kavallieratos avatar
Dimitris Kavallieratos avatar
Mike Cranfield:
Nice work Dimitris - this is a great looking image already!

My thoughts, FWIW,  are that there may be more you could get from the data without needing more integration time.  It looks to me like there may be some remaining gradients - you have applied DBE which is a great too but can be quite challenging to refine and find the best parameters - in my experience it repays some fiddling to achieve the best result.  There is a fair amount of detail you have captured in the surrounding nebulosity - you have tried using curves to introduce contrast, have you tried using the Generalised Hyperbolic Stretch script?  This allows you to do some very targeted stretching to bring contrast into specific areas of your image.  Dave Payne and I are presenting GHS on the Astro Imaging Channel tomorrow which will give some background to how best to use it.  I have not used NoiseExterminator but there are some great reports of it and it certainly seems to have done a nice job on your image here.  It may just be my eyes, but it looks like there may be a slight green cast to your image, a dose of SCNR might help clean that up.

Having said all that, there are many far more experienced image processors than me on this forum who I am sure can give you some much more expert pointers to maximise the potential of the data you have.

Good luck and CS

Mike

GHS is incredible I may say, have not used it yet but will redo my image and come back to you! Thank you for bringing my attention to this.

My integration is here, would you be kind to propose a possible mini workflow for GHS? I am trying different things and even though the nebula comes out all my stars are blowned out Although for a first try it does wonders and it have everything that you need (HT,Lightness HT,Sat etc) it is very good! Also, I hope that I bring out real data and not something that's non existent .

My effort after noise reduction :



NoiseXterminator is just...magic.Especially for me being and beginner in all of this and under 30sec ,it is crazy! Previously even with the nearest to convenience noise reduction method for me(EZ Denoise), the script takes about 15-30 min and have inferior results.

Thank you Mike!
Hi Dimitris,

For a Bortle 8 sky this is a good result, but there is far more dust visible in your image than you think! If you look at the histogram, it is too far over to the right and quite narrow, also the image looks a bit washed out. Just for a bit of fun I converted your image to 16 bit and ran it through Starnet++ to remove the stars. I then gave it a quick stretch in Siril, before setting the black point to 25,25,25 in Photoshop levels. I used a point halfway between the nebula and the bottom right corner which worked well. Finally I reduced the stars slightly before adding them back in. It's not perfect, but it gives an idea of what is hiding in there! Sorry for the poor quality but I only had your 8 bit jpeg to work with. You should get far better results with your original data.

Hope this helps.

Cs, Tony.

iris.jpg

Will redo my image, indeed my histogram was a little narrow, trying to fix this. You did a great job btw.Thanks!
andrea tasselli:
Hello !

Throughout galaxy season I have invest a lot of time-due to focal length shortage so there are not good targets for me this time of the year- on the Iris Nebula and with ~12 hours of good data from Bortle 8 ,I have a kind of acceptable result.

Should I invest more time on this target and have a >20 hrs integration time or it is just a loss of time from there on? That dark nebula that surrounds the Iris is barely visible on my integration, how is it possible that people with about the same integration time as me and LP have achieved fantastic results? I shoot with my ASI533 OSC and L-Pro filter with exposures of 180s and 300s.

Thank you

I think it is a hell of the job with the sort of skies you have and you should be proud of yourself.  Adding data is always worth it as it allows to ease on the NR pedal. This said you would want to remove the copyright right in front of the nebula so we can see it! Stars colours need improving too. Otherwise pretty good going.

Thank you Andrea for the encouragement, will invest up to the 20 hrs mark then and see what I have!
Dale Penkala avatar
Dale Penkala avatar
I have to agree with @Mike Cranfield & @andrea tasselli Demetris! Very nice image! I’m far from an expert but if it were me I’d probable use star exterminator and work on just the nebulosity, maybe some color adjustment or possibly some PCC and maybe use a sharpened/contrast luminance layer to add back a tad more detail and contrast. Maybe a tad bit of star reduction and add the stars back in with pixel math. Just my thoughts. If your ok with it I may download your image and play with it myself and repost here in your thread?

Nice looking image!

Dale
Helpful Respectful Engaging Supportive
Dimitris Kavallieratos avatar
Dimitris Kavallieratos avatar
Dale Penkala:
I have to agree with @Mike Cranfield & @andrea tasselli Demetris! Very nice image! I’m far from an expert but if it were me I’d probable use star exterminator and work on just the nebulosity, maybe some color adjustment or possibly some PCC and maybe use a sharpened/contrast luminance layer to add back a tad more detail and contrast. Maybe a tad bit of star reduction and add the stars back in with pixel math. Just my thoughts. If your ok with it I may download your image and play with it myself and repost here in your thread?

Nice looking image!

Dale


Thank you Dale! Of course you can ,that is the whole point. I added a link to my PI integration on my post above!
Dale Penkala avatar
Dale Penkala avatar
Dale Penkala:
I have to agree with @Mike Cranfield & @andrea tasselli Demetris! Very nice image! I’m far from an expert but if it were me I’d probable use star exterminator and work on just the nebulosity, maybe some color adjustment or possibly some PCC and maybe use a sharpened/contrast luminance layer to add back a tad more detail and contrast. Maybe a tad bit of star reduction and add the stars back in with pixel math. Just my thoughts. If your ok with it I may download your image and play with it myself and repost here in your thread?

Nice looking image!

Dale


Thank you Dale! Of course you can ,that is the whole point. I added a link to my PI integration on my post above!

I have some errands to do but I’ll try to play with it later on this afternoon and post if I like the results I come up with.

Dale
Andy Wray avatar
Andy Wray avatar
I had a quick play and ended up with the below.  I've no idea if it is supposed to look like this, but this is what I got as a beginner.  I'm sure I've over-processed it as I was just playing, but it does show that you have lots of detail in your image.  I think you could end up with a stunning image if you spend some time on it (certainly more than my 10 minutes).  I'm actually not sure that you need any more integration time.  With careful processing I think you have a really good image there.


FWIW:  the nasty, quick and dirty workflow on this was:

* Extract RGB channels
* Automatic background extraction on the 3 channels
* Linear Fit colour calibration with red as the master
* ChannelCombination of 3 channels
* SCNR green noise reduction
* StarNet starmask creation and star removal
* Curves to increase saturation on starless image and darken background
* Pixelmath combination of starless image with starmask
* EZStarReduction
* NoiseXterminator
Helpful Respectful Engaging Supportive
Dale Penkala avatar
Dale Penkala avatar
Hello !

Throughout galaxy season I have invest a lot of time-due to focal length shortage so there are not good targets for me this time of the year- on the Iris Nebula and with ~12 hours of good data from Bortle 8 ,I have a kind of acceptable result.

Should I invest more time on this target and have a >20 hrs integration time or it is just a loss of time from there on? That dark nebula that surrounds the Iris is barely visible on my integration, how is it possible that people with about the same integration time as me and LP have achieved fantastic results? I shoot with my ASI533 OSC and L-Pro filter with exposures of 180s and 300s.

Thank you

Hello Dimitris,
As per earlier, I did 3 different renditions of your image. Keep in mind, the image I downloaded was a Jpeg file. We usually work with Tiff's or Fits files. That said, here is what I did with your image.

1) Ran a DBE just to go through and remove any gradient. I also run some SCNR to remove a tad bit of the green in the image. I think it made a nicer color rendition.
2) Run StarExterminator. Pulled luminance and added a sharpening filter and dropped the background for contrast and added it back to the original image using LRGB Combination. In hindsight, I should have used a range mask to hold back the color to the dusky background, but I made a saturation adjustment on the second image.
3) Took the starless image and run it through Topaz AI and used the "low light" filter.
4) Pulled that back into PI and used HDR Multi Tran to pull out more of the core details. Added the stars back in with pixel math at .75. I was very surprised at how much detail was in that core, which is why I cropped the image and used it for the 3rd rendition. In the Cropped version, I run a very slight star reduction to it as well.
5) Took the final image into Gimp and made some final tweaks to the contrast and color balance.

I hope you like what I did. I'm far from a master at processing, and remember this is "my personal" taste as to the color and brightness. I'm still not real keen on how red the dust clouds are, which is why I did a slight desaturation in the 2nd & 3rd image.

I'm sure many on AB could do a better job, but this was a fun image to work with, and I'll say it again, GREAT JOB ON THE IMAGE!

Dale




Helpful Supportive
Dale Penkala avatar
Dale Penkala avatar
I decided to do another slight star reduction in the cropped version. I think the dust clouds show a bit better.

Dimitris Kavallieratos avatar
Dimitris Kavallieratos avatar
Dale Penkala:
Hello !

Throughout galaxy season I have invest a lot of time-due to focal length shortage so there are not good targets for me this time of the year- on the Iris Nebula and with ~12 hours of good data from Bortle 8 ,I have a kind of acceptable result.

Should I invest more time on this target and have a >20 hrs integration time or it is just a loss of time from there on? That dark nebula that surrounds the Iris is barely visible on my integration, how is it possible that people with about the same integration time as me and LP have achieved fantastic results? I shoot with my ASI533 OSC and L-Pro filter with exposures of 180s and 300s.

Thank you

Hello Dimitris,
As per earlier, I did 3 different renditions of your image. Keep in mind, the image I downloaded was a Jpeg file. We usually work with Tiff's or Fits files. That said, here is what I did with your image.

1) Ran a DBE just to go through and remove any gradient. I also run some SCNR to remove a tad bit of the green in the image. I think it made a nicer color rendition.
2) Run StarExterminator. Pulled luminance and added a sharpening filter and dropped the background for contrast and added it back to the original image using LRGB Combination. In hindsight, I should have used a range mask to hold back the color to the dusky background, but I made a saturation adjustment on the second image.
3) Took the starless image and run it through Topaz AI and used the "low light" filter.
4) Pulled that back into PI and used HDR Multi Tran to pull out more of the core details. Added the stars back in with pixel math at .75. I was very surprised at how much detail was in that core, which is why I cropped the image and used it for the 3rd rendition. In the Cropped version, I run a very slight star reduction to it as well.
5) Took the final image into Gimp and made some final tweaks to the contrast and color balance.

I hope you like what I did. I'm far from a master at processing, and remember this is "my personal" taste as to the color and brightness. I'm still not real keen on how red the dust clouds are, which is why I did a slight desaturation in the 2nd & 3rd image.

I'm sure many on AB could do a better job, but this was a fun image to work with, and I'll say it again, GREAT JOB ON THE IMAGE!

Dale






Great job , definitely better than mine!!

My PI integration original file has been posted above! Here it is if you still have the courage

Anyway, thank you very much for your input and ideas, I will try to reprocess my image incorporating some of your methods!
Stuart Taylor avatar
Stuart Taylor avatar
Interesting thread. Nice image Dimitris! And it really shows how much more can often be made of a dataset that we less experienced people thought was disappointing. I am new to PI and on a steep learning curve, but when I get home from work, I'd like to have a go at your integrated image too.

One question - many people mentioned Star XTerminator. I'd not heard of that, so googled it. It seems to do the same as Starnet/Starnet2 which I use a lot in PI. Is Star XTerminator better than that then?
Well Written Respectful Engaging
Dale Penkala avatar
Dale Penkala avatar
Stuart Taylor:
Interesting thread. Nice image Dimitris! And it really shows how much more can often be made of a dataset that we less experienced people thought was disappointing. I am new to PI and on a steep learning curve, but when I get home from work, I'd like to have a go at your integrated image too.

One question - many people mentioned Star XTerminator. I'd not heard of that, so googled it. It seems to do the same as Starnet/Starnet2 which I use a lot in PI. Is Star XTerminator better than that then?

Hello Stuart,
Star eXterminator is a processing plugin for PI. It isn’t free but certainly works much better then StarNet 2. I didn’t believe it until I tryed the 30 day trail version and ended up buying it.

Dale
Concise
Stuart Taylor avatar
Stuart Taylor avatar
Dale Penkala:
Stuart Taylor:
Interesting thread. Nice image Dimitris! And it really shows how much more can often be made of a dataset that we less experienced people thought was disappointing. I am new to PI and on a steep learning curve, but when I get home from work, I'd like to have a go at your integrated image too.

One question - many people mentioned Star XTerminator. I'd not heard of that, so googled it. It seems to do the same as Starnet/Starnet2 which I use a lot in PI. Is Star XTerminator better than that then?

Hello Stuart,
Star eXterminator is a processing plugin for PI. It isn’t free but certainly works much better then StarNet 2. I didn’t believe it until I tryed the 30 day trail version and ended up buying it.

Dale

Really good to know! Thanks
Stuart Taylor avatar
Stuart Taylor avatar
Stuart Taylor:
Dale Penkala:
Stuart Taylor:
Interesting thread. Nice image Dimitris! And it really shows how much more can often be made of a dataset that we less experienced people thought was disappointing. I am new to PI and on a steep learning curve, but when I get home from work, I'd like to have a go at your integrated image too.

One question - many people mentioned Star XTerminator. I'd not heard of that, so googled it. It seems to do the same as Starnet/Starnet2 which I use a lot in PI. Is Star XTerminator better than that then?

Hello Stuart,
Star eXterminator is a processing plugin for PI. It isn’t free but certainly works much better then StarNet 2. I didn’t believe it until I tryed the 30 day trail version and ended up buying it.

Dale

Really good to know! Thanks

Purchased! Thanks again!