|
Robert Chozick <rchozick@...>
I do the same thing. It’s sometimes works like deconvolution to help later sharpening. The DeNoise software is incredible on any images that have a lot of noise.
Robert
toggle quoted message
Show quoted text
On Jul 27, 2021, at 1:03 PM, Dale Ghent <daleg@...> wrote:
I've played around with Topaz. I got acceptable results by not giving it free reign. I'd make an original and a Topaz'd layer in PS and brush in the Topaz'd layer with varying levels of opacity to bring out its effects in a more subtle manner and in the places I thought it would work the best. I found it a good way to give a hint of sharpened structure to the eye without it being too overbearing as the pure Topaz'd layer would feel.
On Jul 27, 2021, at 13:57, Brian Valente <bvalente@...> wrote:
I am - i'm a longtime user of topaz and topaz AI.
They can be really good, but generally for me they tend to be a bit heavy-handed, especially the denoise algorithms.
I find Topaz clarity and detail are much better for teasing out details.
My hope is Topaz will start their AI training on astro images, which will make their algorithms much better suited for our work.
On Tue, Jul 27, 2021 at 10:01 AM Robert Chozick via groups.io <rchozick@...> wrote:
Are you familiar with Topaz noise reduction and sharpening? It works wonders on some things, especially the DeNoise. It is available standalone or as a PS plugin.
Robert
On Jul 27, 2021, at 11:46 AM, Brian Valente <bvalente@...> wrote:
Yeah it's interesting to see how processing techniques are evolving. Starnet was a curiosity a couple years ago, and now it's front and center!
Brian
On Tue, Jul 27, 2021 at 9:40 AM Robert Chozick via groups.io <rchozick@...> wrote:
I agree. I meant that my method of masking out the target for processing, leaving the stars and background alone, does not work on extended nebula and large galaxies. I am excited for the possibilities of the new method.
Really nice image but the stars are not as pronounced as they would be. I will play with the process.
Robert
On Jul 27, 2021, at 11:15 AM, Brian Valente <bvalente@...> wrote:
Hi Robert
For extended nebula or large galaxies this does not work.
i think it works exceptionally well for extended nebulas, where enhancing the nebulosity without hurting the stars can really bring out structures
Although this image could certainly be improved, I used this approach to pull out nebulosity and blend back with the "regular version". IIRC i did this going between PS and Pix
https://www.brianvalentephotography.com/astrophotography/i-NJN9nHb/A
(can't speak to galaxies because I use other techniques for those)
On Tue, Jul 27, 2021 at 9:09 AM Robert Chozick via groups.io <rchozick@...> wrote:
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work.
Robert
On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote: Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
--
Brian
Brian Valente
portfolio brianvalentephotography.com
--
Brian
Brian Valente
portfolio brianvalentephotography.com
--
Brian
Brian Valente
portfolio brianvalentephotography.com
|
|
|
I've played around with Topaz. I got acceptable results by not giving it free reign. I'd make an original and a Topaz'd layer in PS and brush in the Topaz'd layer with varying levels of opacity to bring out its effects in a more subtle manner and in the places I thought it would work the best. I found it a good way to give a hint of sharpened structure to the eye without it being too overbearing as the pure Topaz'd layer would feel.
toggle quoted message
Show quoted text
On Jul 27, 2021, at 13:57, Brian Valente <bvalente@...> wrote:
I am - i'm a longtime user of topaz and topaz AI.
They can be really good, but generally for me they tend to be a bit heavy-handed, especially the denoise algorithms.
I find Topaz clarity and detail are much better for teasing out details.
My hope is Topaz will start their AI training on astro images, which will make their algorithms much better suited for our work.
On Tue, Jul 27, 2021 at 10:01 AM Robert Chozick via groups.io <rchozick@...> wrote:
Are you familiar with Topaz noise reduction and sharpening? It works wonders on some things, especially the DeNoise. It is available standalone or as a PS plugin.
Robert
On Jul 27, 2021, at 11:46 AM, Brian Valente <bvalente@...> wrote:
Yeah it's interesting to see how processing techniques are evolving. Starnet was a curiosity a couple years ago, and now it's front and center!
Brian
On Tue, Jul 27, 2021 at 9:40 AM Robert Chozick via groups.io <rchozick@...> wrote:
I agree. I meant that my method of masking out the target for processing, leaving the stars and background alone, does not work on extended nebula and large galaxies. I am excited for the possibilities of the new method.
Really nice image but the stars are not as pronounced as they would be. I will play with the process.
Robert
On Jul 27, 2021, at 11:15 AM, Brian Valente <bvalente@...> wrote:
Hi Robert
For extended nebula or large galaxies this does not work.
i think it works exceptionally well for extended nebulas, where enhancing the nebulosity without hurting the stars can really bring out structures
Although this image could certainly be improved, I used this approach to pull out nebulosity and blend back with the "regular version". IIRC i did this going between PS and Pix
https://www.brianvalentephotography.com/astrophotography/i-NJN9nHb/A
(can't speak to galaxies because I use other techniques for those)
On Tue, Jul 27, 2021 at 9:09 AM Robert Chozick via groups.io <rchozick@...> wrote:
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work.
Robert
On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote:
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
--
Brian
Brian Valente
portfolio brianvalentephotography.com
--
Brian
Brian Valente
portfolio brianvalentephotography.com
--
Brian
Brian Valente
portfolio brianvalentephotography.com
|
|
|
I am - i'm a longtime user of topaz and topaz AI.
They can be really good, but generally for me they tend to be a bit heavy-handed, especially the denoise algorithms.
I find Topaz clarity and detail are much better for teasing out details.
My hope is Topaz will start their AI training on astro images, which will make their algorithms much better suited for our work.
Are you familiar with Topaz noise reduction and sharpening? It works wonders on some things, especially the DeNoise. It is available standalone or as a PS plugin.
Robert On Jul 27, 2021, at 11:46 AM, Brian Valente < bvalente@...> wrote: Yeah it's interesting to see how processing techniques are evolving. Starnet was a curiosity a couple years ago, and now it's front and center!
Brian I agree. I meant that my method of masking out the target for processing, leaving the stars and background alone, does not work on extended nebula and large galaxies. I am excited for the possibilities of the new method.
Really nice image but the stars are not as pronounced as they would be. I will play with the process.
Robert On Jul 27, 2021, at 11:15 AM, Brian Valente < bvalente@...> wrote: Hi Robert
>>>For extended nebula or large galaxies this does not work.
i think it works exceptionally well for extended nebulas, where enhancing the nebulosity without hurting the stars can really bring out structures
Although this image could certainly be improved, I used this approach to pull out nebulosity and blend back with the "regular version". IIRC i did this going between PS and Pix
(can't speak to galaxies because I use other techniques for those)
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work. On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote:
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
--
--
|
|
|
Robert Chozick <rchozick@...>
Are you familiar with Topaz noise reduction and sharpening? It works wonders on some things, especially the DeNoise. It is available standalone or as a PS plugin.
On Jul 27, 2021, at 11:46 AM, Brian Valente < bvalente@...> wrote: Yeah it's interesting to see how processing techniques are evolving. Starnet was a curiosity a couple years ago, and now it's front and center!
Brian I agree. I meant that my method of masking out the target for processing, leaving the stars and background alone, does not work on extended nebula and large galaxies. I am excited for the possibilities of the new method.
Really nice image but the stars are not as pronounced as they would be. I will play with the process.
Robert On Jul 27, 2021, at 11:15 AM, Brian Valente < bvalente@...> wrote: Hi Robert
>>>For extended nebula or large galaxies this does not work.
i think it works exceptionally well for extended nebulas, where enhancing the nebulosity without hurting the stars can really bring out structures
Although this image could certainly be improved, I used this approach to pull out nebulosity and blend back with the "regular version". IIRC i did this going between PS and Pix
(can't speak to galaxies because I use other techniques for those)
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work. On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote:
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
--
--
|
|
|
Yeah it's interesting to see how processing techniques are evolving. Starnet was a curiosity a couple years ago, and now it's front and center!
Brian
I agree. I meant that my method of masking out the target for processing, leaving the stars and background alone, does not work on extended nebula and large galaxies. I am excited for the possibilities of the new method.
Really nice image but the stars are not as pronounced as they would be. I will play with the process.
Robert On Jul 27, 2021, at 11:15 AM, Brian Valente < bvalente@...> wrote: Hi Robert
>>>For extended nebula or large galaxies this does not work.
i think it works exceptionally well for extended nebulas, where enhancing the nebulosity without hurting the stars can really bring out structures
Although this image could certainly be improved, I used this approach to pull out nebulosity and blend back with the "regular version". IIRC i did this going between PS and Pix
(can't speak to galaxies because I use other techniques for those)
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work. On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote:
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
--
|
|
|
Robert Chozick <rchozick@...>
I agree. I meant that my method of masking out the target for processing, leaving the stars and background alone, does not work on extended nebula and large galaxies. I am excited for the possibilities of the new method.
Really nice image but the stars are not as pronounced as they would be. I will play with the process.
On Jul 27, 2021, at 11:15 AM, Brian Valente < bvalente@...> wrote: Hi Robert
>>>For extended nebula or large galaxies this does not work.
i think it works exceptionally well for extended nebulas, where enhancing the nebulosity without hurting the stars can really bring out structures
Although this image could certainly be improved, I used this approach to pull out nebulosity and blend back with the "regular version". IIRC i did this going between PS and Pix
(can't speak to galaxies because I use other techniques for those)
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work. On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote:
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
--
|
|
|
Hi Robert
>>>For extended nebula or large galaxies this does not work.
i think it works exceptionally well for extended nebulas, where enhancing the nebulosity without hurting the stars can really bring out structures
Although this image could certainly be improved, I used this approach to pull out nebulosity and blend back with the "regular version". IIRC i did this going between PS and Pix
(can't speak to galaxies because I use other techniques for those)
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work. On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote:
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
|
|
|
Robert Chozick <rchozick@...>
Thanks. This sounds like a very useful tool. As you all know, destruction of the stars is so easy with all the processes we do to make our pretty pictures. Around a small target in Photoshop I can bring back the original background and stars easily. For extended nebula or large galaxies this does not work.
On Jul 27, 2021, at 10:23 AM, Brian Valente <bvalente@...> wrote:
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
|
|
|
Starnet only works on stretched images and is now part of regular PixInsight distribution
The "starmask" version produces a color version of the stars, which is good for blending back, but this approach can have some artifacting around the stars. A typical application is to use starnet to create the starless version for enhancing the nebulosity, then blending back into the "regular" image version using something like Screen blend mode (PS or PixInsight). ymmv
Brian
|
|
|
Robert Chozick <rchozick@...>
Thanks Glenn. I will check out the application. I’ve been lazy and not downloaded the newer version of Pixinsight. Wouldn’t the star mask only be a gray representation of where the stars were and not the actual image of the stars themselves?
On Jul 27, 2021, at 9:43 AM, Glenn <public@...> wrote:
I forgot to mention that once you have a starless image, you can import it into Photoshop and do whatever you want with it.
Best,
Glenn
|
|
|
I forgot to mention that once you have a starless image, you can import it into Photoshop and do whatever you want with it.
Best,
Glenn
|
|
|
Thank you, Robert.
The process called Starnet+, which is also integrated into PixInsight, does not automatically produce a separate image with the stars that were just removed. However, doing so is easy and only requires checking the "make star mask" button in the process window. I have been doing it on stretched images but I think the process may be been updated to work on linear ones—you would have to verify. You could make a starless image either in PixInsight or by running Starnet+ as a separate application. The program is available on one of the more popular open-source repositories (can't remember which).
Best,
Glenn
|
|
|
Hello Glenn,
Thanks for sharing! Very nice image.
I too, just recently started to capture the Elephant Trunk Nebula but still need to process the images. I use a quad narrowband filter with a ZWO ASI2600 color camera on a 130GTX. Did stacks at 240, 360, and 720 seconds. Here in California, even with a nearly full moon I got decent results with the filter in place. Now I need to start processing in PixInSight.
Having your image as a reference is helpful.
Best,
Alex
|
|
|
Robert Chozick <rchozick@...>
Great image Glenn.
You now have peaked my interest in this star removal process. I will have to play with it.
Does it place the stars into a separate Image. I am wondering if I can move that star image into Photoshop. I use Pixinsight for a few processes but do most of my processing in Photoshop. I usually only use dynamic background extraction and Photometric Color Calibration. For images at my new dark sky site I found I don’t even need DBE anymore.
On Jul 24, 2021, at 1:13 PM, Glenn <public@...> wrote:
Hi gang, I had a stretch of good weather recently and was able to capture the elephant trunk with narrowband filters over several nights. This is one of three images I was able to do during the first outing with my new 10-inch Newtonian scope. I was not sure if my Mach1 would be able to handle such a heavy and long scope, but it hardly seemed to notice. It is a testament to A-P mount design. For higher resolution and details, please check my AstroBin page. Thanks for looking! Glenn
<IC1396 elephant final 2.jpeg>
|
|
|
Robert Chozick <rchozick@...>
Great image Glenn.
You now have peaked my interest in this star removal process. I will have to play with it.
Does it place the stars into a separate Image. I am wondering if I can move that star image into Photoshop. I use Pixinsight for a few processes but do most of my processing in Photoshop. I usually only use dynamic background extraction and Photometric Color Calibration. For images at my new dark sky site I found I don’t even need DBE anymore.
On Jul 24, 2021, at 1:13 PM, Glenn <public@...> wrote:
Hi gang, I had a stretch of good weather recently and was able to capture the elephant trunk with narrowband filters over several nights. This is one of three images I was able to do during the first outing with my new 10-inch Newtonian scope. I was not sure if my Mach1 would be able to handle such a heavy and long scope, but it hardly seemed to notice. It is a testament to A-P mount design. For higher resolution and details, please check my AstroBin page. Thanks for looking! Glenn
<IC1396 elephant final 2.jpeg>
|
|
|
Karen Christen
That’s a really pleasing image, Glenn. Thanks for sharing it with us.
Karen
AP
toggle quoted message
Show quoted text
From: main@ap-gto.groups.io <main@ap-gto.groups.io>
On Behalf Of Glenn
Sent: Saturday, July 24, 2021 1:14 PM
To: main@ap-gto.groups.io
Subject: [ap-gto] The Elephant Trunk Nebula in the Hubble palette
Hi gang,
I had a stretch of good weather recently and was able to capture the elephant trunk with narrowband filters over several nights. This is one of three images I was able to do during the first outing with my new 10-inch Newtonian scope. I was not sure if my Mach1
would be able to handle such a heavy and long scope, but it hardly seemed to notice. It is a testament to A-P mount design.
For higher resolution and details, please check my
AstroBin page. Thanks for looking!
Glenn
-- Karen ChristenAstro-Physics
|
|
|
M Hambrick
Thanks Glenn and Steve. I will definitely check those out.
Mike
|
|
|
Steve beat me to it. Adam has two outstanding video series: the Fundamentals series for beginner/intermediate and Horizons for more advanced users. I subscribe to both and consider my subscriptions to be some of the best money I've spent in this hobby. When you consider how much an in-person course costs, the price of his videos is a bargain. He must have more than a hundred hours of video in his two offerings. What I love about his instruction is that he does not only teach the how but the why. He does not give the cookbook approach to processing. Check it out! https://adamblockstudios.com/categories/PixInsightBest, Glenn
|
|
|
Mike you may want to look at Adam Block’s series on image processing. It’s extensive and covers a vast majority of processes and is a subscription based service backed with an active forum. -Steve
From: main@ap-gto.groups.io <main@ap-gto.groups.io> On Behalf Of M Hambrick Sent: Sunday, July 25, 2021 8:57 AM To: main@ap-gto.groups.io Subject: Re: [ap-gto] The Elephant Trunk Nebula in the Hubble palette Thanks for sharing your techniques Glen.
I wouldn't describe myself as a beginner, but my processing techniques are nowhere near as sophisticated as yours. I have been doing all of my processing using MaxIm DL Pro, but your comments along with tips I have gotten from others have me convinced that I need to investigate some of the other processing software.
Of course having the software is only the beginning. Then I have to learn how to use it. Even with MaxIm DL I have not come close to exploiting all of the capabilities that it has. It would really be great if I could find an image processing course or seminar to attend at NEAF or one of the other astronomy conferences.
Mike
|
|
|
M Hambrick
Thanks for sharing your techniques Glen.
I wouldn't describe myself as a beginner, but my processing techniques are nowhere near as sophisticated as yours. I have been doing all of my processing using MaxIm DL Pro, but your comments along with tips I have gotten from others have me convinced that I need to investigate some of the other processing software.
Of course having the software is only the beginning. Then I have to learn how to use it. Even with MaxIm DL I have not come close to exploiting all of the capabilities that it has. It would really be great if I could find an image processing course or seminar to attend at NEAF or one of the other astronomy conferences.
Mike
|
|
|
