Hi, this is my longest integration so far, and I really really wanna make the most out of the data. My workflow, due to cone error was like this: 1) stack all subs that were good (639) and do another stack with just the subs before the meridian flip 2) GraXpert on both 3) then in PI on both: Background Neutralisation, Spectrophotometric CC, SCNR, MMT on Chrominance Noise with Mask applied, BlurX, StarX on both. Then I kept working on the starless 20h Nebula and the stars which were only around 10 hours I'd guess. Starless: 4) NoiseX 5) Multiple Stretches with GHS 6) Cloning Image, applied HDRMultiscaleTransform on the Nebula Core 7) Added with HDR Nebula with the regular in Pixel math, weighted 40 HDR, 60 Regular 8) Some Curves Transformations for Saturation / Contrast with Mask 9) Another Round of Background Neutralisation because it was way too blueish / purplish Star-Stack: Stretching and Saturation slightly inscreased. Merged both images with Pixelmath: combine(Starless, Stars, op_screen()) Please give honest feedback, Image can be found here: https://www.astrobin.com/n2apl4/Thank you and Clear Skies!
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
Well, I don't have anything to say… maaaaybe I would be gentler with SCNR? Maybe that would help with the purplish background and would make the image richer in terms of color?
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
I'd control the amount of stretching to tame the noise and also try to suppress the amount of purple-ish fringing you got.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
A quick fix for the magenta colour cast is to invert the image which will turn the magenta green, apply SCNR, then invert the image back again. This will give you a nice neutral result to work with. Another thing I'd experiment with is Contrast Limited Adaptive Histogram Equalisation or whatever it's called in Pixinsight, just a tiny amount will improve the contrast and enhance the details. Hope this helps!
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
I use LocalHistogramEqualization (LHE) to enhance dark gas detail.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
A quick fix for the magenta colour cast is to invert the image which will turn the magenta green, apply SCNR, then invert the image back again. This will give you a nice neutral result to work with. Another thing I'd experiment with is Contrast Limited Adaptive Histogram Equalisation or whatever it's called in Pixinsight, just a tiny amount will improve the contrast and enhance the details. Hope this helps! Hey thanks, should I invert the image in the linear stage? I use LocalHistogramEqualization (LHE) to enhance dark gas detail. Thanks - definitely a good idea! andrea tasselli:
I'd control the amount of stretching to tame the noise and also try to suppress the amount of purple-ish fringing you got. I don‘t think the noise is that bad? The stars are kinda meh, yeah, I might stretch them a little bit less. I only recently started to stretch stars seperately, and it‘s still tricky to figure out the right amount.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
If you find it difficult to stretch stars you could try the scripts from SetiAstro. The star stretching script works pretty good.
Not sure if it's been mentioned but I noticed some areas around the edges that ought to be cropped as they show signs of stacking artifacts, I'd adress that first of all as this can throw off processes like background extraction. Other than that I don't have too much to add to the conversation that hasn't already been mentioned.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
Mina B.:
Hey thanks, should I invert the image in the linear stage? Try it on your fully processed image, it should make a big difference. This technique can also be used to remove magenta rings from around stars, which is a common problem.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
GraXpert is cool and all, but in a dusty situation like this you will want to judiciously apply DBE yourself. By "judiciously" I mean fine tuning the locations of your DBE points will be one of the largest parts of your processing flow.
You can hunt down images with crazy integration times to get an idea of where there is and where there is not dust in your FoV. You can also run DBE on your image, and based on the results go back to the original and keep fine tuning the points.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
GraXpert is cool and all, but in a dusty situation like this you will want to judiciously apply DBE yourself. By "judiciously" I mean fine tuning the locations of your DBE points will be one of the largest parts of your processing flow.
You can hunt down images with crazy integration times to get an idea of where there is and where there is not dust in your FoV. You can also run DBE on your image, and based on the results go back to the original and keep fine tuning the points. I absolutely second this. After cropping DBE will run circles around GraXpert. More than once have I noticed GraXpert obliterate fine details, colors and creating imbalanced backgrounds. If I ever need to run a more automatic process I would much rather use ABE or the atuomatic DBE script from SetiAstro.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
If you find it difficult to stretch stars you could try the scripts from SetiAstro. The star stretching script works pretty good.
Not sure if it's been mentioned but I noticed some areas around the edges that ought to be cropped as they show signs of stacking artifacts, I'd adress that first of all as this can throw off processes like background extraction. Other than that I don't have too much to add to the conversation that hasn't already been mentioned. Downloaded the script, gonna give it a try, thanks! And yep, I also saw upon pixel peeping - I cropped the corners a tad bit more now. Mina B.:
Hey thanks, should I invert the image in the linear stage?
Try it on your fully processed image, it should make a big difference. This technique can also be used to remove magenta rings from around stars, which is a common problem.
Thanks, will do in the end! GraXpert is cool and all, but in a dusty situation like this you will want to judiciously apply DBE yourself. By "judiciously" I mean fine tuning the locations of your DBE points will be one of the largest parts of your processing flow.
You can hunt down images with crazy integration times to get an idea of where there is and where there is not dust in your FoV. You can also run DBE on your image, and based on the results go back to the original and keep fine tuning the points.
I absolutely second this. After cropping DBE will run circles around GraXpert. More than once have I noticed GraXpert obliterate fine details, colors and creating imbalanced backgrounds. If I ever need to run a more automatic process I would much rather use ABE or the atuomatic DBE script from SetiAstro.
Yeah, I tried the ADBE script now, it looks a lot better already than GraxPert. basically I have mostly Nebula in my pretty small FOV, there isn't much actual background but GraXpert basically killed off the fainter background completely.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
Mina B.:
Hi, this is my longest integration so far, and I really really wanna make the most out of the data.
My workflow, due to cone error was like this:
1) stack all subs that were good (639) and do another stack with just the subs before the meridian flip
2) GraXpert on both
3) then in PI on both: Background Neutralisation, Spectrophotometric CC, SCNR, MMT on Chrominance Noise with Mask applied, BlurX, StarX on both.
Then I kept working on the starless 20h Nebula and the stars which were only around 10 hours I'd guess.
Starless:
4) NoiseX
5) Multiple Stretches with GHS
6) Cloning Image, applied HDRMultiscaleTransform on the Nebula Core
7) Added with HDR Nebula with the regular in Pixel math, weighted 40 HDR, 60 Regular
8) Some Curves Transformations for Saturation / Contrast with Mask
9) Another Round of Background Neutralisation because it was way too blueish / purplish
Star-Stack:
Stretching and Saturation slightly inscreased.
Merged both images with Pixelmath: combine(Starless, Stars, op_screen())
Please give honest feedback, Image can be found here: https://www.astrobin.com/n2apl4/
Thank you and Clear Skies! There's nothing that can be critiqued, certainly not by anyone who has not done comparable work––something I check when receiving advice. One of my mentors used to say. "If you don't know, don't help me." It is an exceptional image technically and stylistically! Well done!
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
I think the overall result is fine, albeit you can do some adjustments like other people is mentioning. However, there are some things you could change in your work flow that I am sure it will help:
1) For heavily filled FOV like yours, I would ditch any automated tool to remove gradients and would use DBE. You have several spots to place samples which should deal with the gradients (assuming they are not horrible) - remember that some regions might have nebulosity but it is so faint that you cannot make anything useful out of it and therefore should be treated like background;
2) Create a synthetic luminance and follow an LRGB workflow. It will allow you to get less noise and better contrast right from the onset and you'll still be able to improve it further;
3) Use HistogramTransformation plus HDRMultiscale instead of GHS. I think GHS overly complicated things and usually I don't like the lack of contrast it provides - HT provides by far the most pleasing contrast to my eyes and if a region is blown out, HDRMultiscale usually solves it. But for it, choose a Gaussian function instead of the default.
The other processes are a bit personal taste but these technical ones should help you having a better image to work on.
I hope this helps. 😊
Andre
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
Andre Vilhena:
I think the overall result is fine, albeit you can do some adjustments like other people is mentioning. However, there are some things you could change in your work flow that I am sure it will help:
1) For heavily filled FOV like yours, I would ditch any automated tool to remove gradients and would use DBE. You have several spots to place samples which should deal with the gradients (assuming they are not horrible) - remember that some regions might have nebulosity but it is so faint that you cannot make anything useful out of it and therefore should be treated like background; On this one - I realised my FOV is so filled, I barely have any background. I'm not even sure if I have a gradient or if I can ditch it completely... There was no moon up while imaging, there wasn't any light pollution because I was at a star party under Bortle 3 skies and people were super considerate of us photographers. ADBE kinda cooked the Dark Nebula, I realised in non-linear state, it looked totally fried in the end.. and GraXpert wasn't the best probably but I still prefer my first try. Now I try DBE, but I'm not sure if it will do the trick... I only placed boxes on where real sky background is, not any dark nebula, and there weren't many spots... I think it probably introduced a gradient were there was barely any. I give it a try, if it looks not good, I'm gonna process it without doing anything to the background for once.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
BryanHudson:
There's nothing that can be critiqued, certainly not by anyone who has not done comparable work––something I check when receiving advice. One of my mentors used to say. "If you don't know, don't help me."
It is an exceptional image technically and stylistically! Well done! Oh and thanks, that's very kind! But I'm sadly never quite content with an image. I'm glad for all the helpful tips I got here though.
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
Mina B.:
under Bortle 3 skies This is important piece of information. Likely, at worst, you may have only slight gradients. I find easier to check in the monochrome image so probably you can have a look at the red and blue channels to see if you identify any gradient. If I were to use DBE, I would place the samples in the place the red dots are and set smoothness to something above 0.5 or 0.6. Check the gradient to make sure it is really smooth.  |
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
@Andre Vilhena thank you for the indepth explanation, that really helps! I uploaded a version I'm content with now, it's public - thanks for everyone that helped with tips, and CC. Clear Skies!
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
Mina B.:
BryanHudson:
There's nothing that can be critiqued, certainly not by anyone who has not done comparable work––something I check when receiving advice. One of my mentors used to say. "If you don't know, don't help me."
It is an exceptional image technically and stylistically! Well done!
Oh and thanks, that's very kind! But I'm sadly never quite content with an image. I'm glad for all the helpful tips I got here though.
Yes, the nature of this hobby is that we are never satisfied!
That is likely a little bit of motivating OCD. :-) |
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
Mina B.:
Hey thanks, should I invert the image in the linear stage?
Try it on your fully processed image, it should make a big difference. This technique can also be used to remove magenta rings from around stars, which is a common problem.
The easiest way to do this is using the “CorrectMagentaStars” script in PI. I like to use this myself! Dale
|
You cannot like this item. Reason: "ANONYMOUS".
You cannot remove your like from this item.
Editing a post is only allowed within 24 hours after creating it.
You cannot Like this post because the topic is closed.
Copy the URL below to share a direct link to this post.
This post cannot be edited using the classic forums editor.
To edit this post, please enable the "New forums experience" in your settings.
