Captured my first Mono data sets ( broadband and Narrowband) last week using my new 2600MM ( I’ve been using my 2600MC OSC for 3 years and thought it was time to take the plunge into Mono )
I’m having lots of issues with Deep Sky Stacker at the moment as when I register and stack say my Ha set and load into Compose , my initial AutoDev image has a tiny faint grid pattern across the whole background.
At first I thought it was my capture software APT doing something funky with assigning Bayered fits headers , but check that and the files are good , definitely mono fits
A member on our local forum suggested to try another stacking program like ASTAP.
Ive used ASTAP for just reviewing subs but not stacking
Can someone advise how to stack say a Ha set and an Oiii set and ensure they are aligned ready to load into Compose
I’ve only just fiddled around with ASTAP in the past day learning how to stack one data set like Ha
Also can you combine 2 different nights data , say Ha , as one stack or is there a better way ( DSS uses groups and reference frame to do this )
Any help would be greatly appreciated
Cheers
Martin
A frustrated Astrophotographer
ASTAP and Startools - Mono Imaging Help !
ASTAP and Startools - Mono Imaging Help !
- Attachments
-
-
- IMG_4615.jpeg (170.09 KiB) Viewed 21941 times
-
-
Mike in Rancho
- Posts: 1166
- Joined: Sun Jun 20, 2021 10:05 pm
- Location: Alta Loma, CA
Re: ASTAP and Startools - Mono Imaging Help !
Hello there Frustrated Martin. 
Well, mono, and especially once you start getting fancy like LRGBHa, is where PI's WBPP really shines. But, that cost of entry, especially if you're not going to be using PI for any post processing. Both Freddy and I are doing it though.
I have no experience with APP but it is supposedly an excellent stacker too. Siril may also have the capabilities that are needed here, but I just still find that program thoroughly confounding. If you can figure it out though, it might do the trick.
That grid pattern is a bit interesting, especially since it appears to be warped. Splined by interpolation maybe? In any event, I wonder what the source is?
Since you already have a 2600, you probably already know that its darks and bias, when hyper hyper stretched, will reveal vertical and horizontal banding. But it is at such low levels that normal stretching should never bring it out. I wonder then about your acquisition settings and calibration files. Are you using something like Gain 0 Offset 30 for everything, or maybe that for broadband and Gain 100 Offset 30 (or 50) for Narrowband, with appropriate matching settings for the corresponding darks, flats, and bias? All files taken through the same program, APT? Everything set up correctly in DSS? And by that I mean no reusage of master calibration files, unless you are totally sure you know what you are loading in, and that the digital development process settings are correct for mono (you have to uncheck the top box)?
Moving on to ASTAP, it's a handy stacker, but like DSS has it's own way of doing things and pros/cons to deal with.
First, unlike DSS, you can't just load in Bias (hopefully you are using darks anyway) and have ASTAP know to use it on both sides, lights and flats. You'll have to load files into all four slots. But, bias of course can be loaded in lieu of dark flats, and while not perfect those bias can also be loaded into darks. It will create a master dark, which you can archive and load in by itself in the future (if applicable) to avoid recombination.
I always tended to do things step by step rather than just load and hit go, so I could inspect what was happening.
So you can load your bias into dark flats and hit analyze if you want. Then same with your flats. In the flats tab, you can hit "replace with" and it will combine the bias and use that to calibrate the combined flats, and you will end up with a master. All other files in the flats and dark flats tabs will be disappeared. If you select Classify By Filter you will get masters for each filter, which of course you want.
That actually gets set up in the Stack Method tab. There, make sure that your naming conventions for the filters match between ASTAP and APT (or NINA). You will know you got that right because when you analyze frames that have filters (lights, flats), the filter column will change color to indicate that it is recognized. Also for mono make sure that nothing is checked in the RAW/OSC settings. Lastly in Stack Method, you are going to want to choose Sigma Clip, skip LRGB combine. This will then create multiple light filter masters (albeit not registered, more on that later).
Then you can load your darks, which can be analyzed and/or replaced with master. Darks also (and Flats) have options across the top for Classify by things like date, which permits the session (group) stacking as you mentioned.
Finally you can load your lights. That must be followed by analyze/organize, which is very similar to running pre-registration in DSS. Once you are happy with the subs you want to stack, you'll probably go set your alignment method tab to star aligned, and then you can just hit stack and watch it go.
If all works well, you will end up with a stack for each filter. "Masterlight" in PI lingo. But, they are not star aligned to each other. Unlike with DSS, you cannot select a reference. ASTAP does that for you, based primarily on HFR/scoring, and that's a drawback for mono, as you can't freshly stack everything to a reference.
So, you clear everything out and then load those master stacks into lights. You may have to manually check the boxes. Then run analyze. Then in Stack Method you would choose "calibration and alignment only." Don't worry about calibration since you have already done that (but again make sure you have cleared out the files from the darks and flats tabs). Then run "stack" and it should then star align all the files to whichever stack had the best HFR, and will spit out new ones - the filename will identify the aligned files.
And those you should be able to then load into compose in ST. However, I can't remember if ST 1.9 is able to pull in the exposure times per filter from ASTAP after having to run that separate star alignment (another benefit of WBPP).
Well, now that I've thoroughly confused you, just go play with it. It's "close enough" in UI to DSS that some experimentation will get you moving along in short order. Also, the ASTAP website has pretty decent instructions.
Well, mono, and especially once you start getting fancy like LRGBHa, is where PI's WBPP really shines. But, that cost of entry, especially if you're not going to be using PI for any post processing. Both Freddy and I are doing it though.
I have no experience with APP but it is supposedly an excellent stacker too. Siril may also have the capabilities that are needed here, but I just still find that program thoroughly confounding. If you can figure it out though, it might do the trick.
That grid pattern is a bit interesting, especially since it appears to be warped. Splined by interpolation maybe? In any event, I wonder what the source is?
Since you already have a 2600, you probably already know that its darks and bias, when hyper hyper stretched, will reveal vertical and horizontal banding. But it is at such low levels that normal stretching should never bring it out. I wonder then about your acquisition settings and calibration files. Are you using something like Gain 0 Offset 30 for everything, or maybe that for broadband and Gain 100 Offset 30 (or 50) for Narrowband, with appropriate matching settings for the corresponding darks, flats, and bias? All files taken through the same program, APT? Everything set up correctly in DSS? And by that I mean no reusage of master calibration files, unless you are totally sure you know what you are loading in, and that the digital development process settings are correct for mono (you have to uncheck the top box)?
Moving on to ASTAP, it's a handy stacker, but like DSS has it's own way of doing things and pros/cons to deal with.
First, unlike DSS, you can't just load in Bias (hopefully you are using darks anyway) and have ASTAP know to use it on both sides, lights and flats. You'll have to load files into all four slots. But, bias of course can be loaded in lieu of dark flats, and while not perfect those bias can also be loaded into darks. It will create a master dark, which you can archive and load in by itself in the future (if applicable) to avoid recombination.
I always tended to do things step by step rather than just load and hit go, so I could inspect what was happening.
So you can load your bias into dark flats and hit analyze if you want. Then same with your flats. In the flats tab, you can hit "replace with" and it will combine the bias and use that to calibrate the combined flats, and you will end up with a master. All other files in the flats and dark flats tabs will be disappeared. If you select Classify By Filter you will get masters for each filter, which of course you want.
That actually gets set up in the Stack Method tab. There, make sure that your naming conventions for the filters match between ASTAP and APT (or NINA). You will know you got that right because when you analyze frames that have filters (lights, flats), the filter column will change color to indicate that it is recognized. Also for mono make sure that nothing is checked in the RAW/OSC settings. Lastly in Stack Method, you are going to want to choose Sigma Clip, skip LRGB combine. This will then create multiple light filter masters (albeit not registered, more on that later).
Then you can load your darks, which can be analyzed and/or replaced with master. Darks also (and Flats) have options across the top for Classify by things like date, which permits the session (group) stacking as you mentioned.
Finally you can load your lights. That must be followed by analyze/organize, which is very similar to running pre-registration in DSS. Once you are happy with the subs you want to stack, you'll probably go set your alignment method tab to star aligned, and then you can just hit stack and watch it go.
If all works well, you will end up with a stack for each filter. "Masterlight" in PI lingo. But, they are not star aligned to each other. Unlike with DSS, you cannot select a reference. ASTAP does that for you, based primarily on HFR/scoring, and that's a drawback for mono, as you can't freshly stack everything to a reference.
So, you clear everything out and then load those master stacks into lights. You may have to manually check the boxes. Then run analyze. Then in Stack Method you would choose "calibration and alignment only." Don't worry about calibration since you have already done that (but again make sure you have cleared out the files from the darks and flats tabs). Then run "stack" and it should then star align all the files to whichever stack had the best HFR, and will spit out new ones - the filename will identify the aligned files.
And those you should be able to then load into compose in ST. However, I can't remember if ST 1.9 is able to pull in the exposure times per filter from ASTAP after having to run that separate star alignment (another benefit of WBPP).
Well, now that I've thoroughly confused you, just go play with it. It's "close enough" in UI to DSS that some experimentation will get you moving along in short order. Also, the ASTAP website has pretty decent instructions.
Re: ASTAP and Startools - Mono Imaging Help !
Mike,
Thanks you’ve made my day
Followed your ASTAP procedures and stacked just a quick Ha set
Loaded into Startools via Compose , opened it , AutoDev inspect > Bin 60% > Crop > quick AutoDev stretch
Bingo !!
Quick snap of image with iPhone
Nice even flat field , no grid or checkered background ( be it a tad noisy but expected under Bortle 8
Goodbye DSS you have served me well for over 6 years
ASTAP now for all future Mono imaging !!
Now….
Not quite clear on the last section……,
“So, you clear everything out and then load those master stacks into lights. You may have to manually check the boxes. Then run analyze. Then in Stack Method you would choose "calibration and alignment only." Don't worry about calibration since you have already done that (but again make sure you have cleared out the files from the darks and flats tabs). Then run "stack" and it should then star align all the files to whichever stack had the best HFR, and will spit out new ones - the filename will identify the aligned files.”
When you say load those “master stacks” I’m assuming you mean load each filter stack eg: Ha , Oiii ,Sii say all 3 together in Lights tab ??
If you then calibrate and align this combo Ha ,Oiii, Sii stack and then stack them , would this be a jumbled mess ??
Not quite on the page with this final procedure
Please excuse my lack of understanding on this bit
Martin
Thanks you’ve made my day
Followed your ASTAP procedures and stacked just a quick Ha set
Loaded into Startools via Compose , opened it , AutoDev inspect > Bin 60% > Crop > quick AutoDev stretch
Bingo !!
Quick snap of image with iPhone
Nice even flat field , no grid or checkered background ( be it a tad noisy but expected under Bortle 8
Goodbye DSS you have served me well for over 6 years
ASTAP now for all future Mono imaging !!
Now….
Not quite clear on the last section……,
“So, you clear everything out and then load those master stacks into lights. You may have to manually check the boxes. Then run analyze. Then in Stack Method you would choose "calibration and alignment only." Don't worry about calibration since you have already done that (but again make sure you have cleared out the files from the darks and flats tabs). Then run "stack" and it should then star align all the files to whichever stack had the best HFR, and will spit out new ones - the filename will identify the aligned files.”
When you say load those “master stacks” I’m assuming you mean load each filter stack eg: Ha , Oiii ,Sii say all 3 together in Lights tab ??
If you then calibrate and align this combo Ha ,Oiii, Sii stack and then stack them , would this be a jumbled mess ??
Not quite on the page with this final procedure
Please excuse my lack of understanding on this bit
Martin
- Attachments
-
-
- IMG_4652.jpeg (106.02 KiB) Viewed 21929 times
-
-
Mike in Rancho
- Posts: 1166
- Joined: Sun Jun 20, 2021 10:05 pm
- Location: Alta Loma, CA
Re: ASTAP and Startools - Mono Imaging Help !
Hey Martin, great job that looks good! Not too shabby to have success on the first try.
So yes now it gets more complex if you want to throw multiple filters all in the same pot. And the next step will be multiple sessions or groups, but I guess lets just stick to more than one filter for now.
And this is where you will want to start using the Classify By checkboxes too.
But assuming you got past that and ended up with (for a SHO target) three stacks - Sii, Ha, and Oiii - created by ASTAP. Each one, however, will have been stacked to its own best sub reference, and that means each stack cannot yet be composed because they are not star-aligned to each other. So, you have to register these finished stacks.
The way I described is one way to accomplish that in ASTAP. Remove/clear all files from the tabs. Now load those stacks into lights. For SHO of course you will have three files. Because ASTAP will recognize these files as stacks, not subs, they may be unchecked by default, and so you might have to check the boxes (all three). Then choose "cal and align only" in the method, and then press the stack button, or press analyze first and stack after. Three new aligned files will be created. These are the same stacks, just now star-aligned i.e. registered. Despite the name they are not calibrated, first because ASTAP can already read from the headers that they are finished (calibrated) stacks, and also to be safe you have removed all files from the calibration tabs anyway. So they are only aligned. No different than running Star Align on finished stacks in PI, and DSS can pull that trick off too for that matter.
They are not actually "stacked" into one either, just think of it as the GO or RUN button.
A second method of doing the same thing in ASTAP is to load your finished filter stacks (LRGB, or SHO, or all 7!) into the Blink tab. There, tick off the Align Images checkbox. And then over on the far side use the Export Aligned button. The result is the same, you will get new files with "_aligned" at the end of the name. And those are now good for ST compose, though again with a possible caveat about auto-reading the total exposure time correctly.
Method 2 might be more intuitive. Whichever works for ya.
Anyway, it's not necessarily super optimal per Ivo, but since ASTAP does not allow manual override of reference selection, post-stacking registration is the only option I'm aware of (absent, perhaps, running cal and align only on all the subs first, and then stacking those - but I'm not 100% sure on that in the multiple mono filter scenario).
So yes now it gets more complex if you want to throw multiple filters all in the same pot. And the next step will be multiple sessions or groups, but I guess lets just stick to more than one filter for now.
And this is where you will want to start using the Classify By checkboxes too.
But assuming you got past that and ended up with (for a SHO target) three stacks - Sii, Ha, and Oiii - created by ASTAP. Each one, however, will have been stacked to its own best sub reference, and that means each stack cannot yet be composed because they are not star-aligned to each other. So, you have to register these finished stacks.
The way I described is one way to accomplish that in ASTAP. Remove/clear all files from the tabs. Now load those stacks into lights. For SHO of course you will have three files. Because ASTAP will recognize these files as stacks, not subs, they may be unchecked by default, and so you might have to check the boxes (all three). Then choose "cal and align only" in the method, and then press the stack button, or press analyze first and stack after. Three new aligned files will be created. These are the same stacks, just now star-aligned i.e. registered. Despite the name they are not calibrated, first because ASTAP can already read from the headers that they are finished (calibrated) stacks, and also to be safe you have removed all files from the calibration tabs anyway. So they are only aligned. No different than running Star Align on finished stacks in PI, and DSS can pull that trick off too for that matter.
They are not actually "stacked" into one either, just think of it as the GO or RUN button.
A second method of doing the same thing in ASTAP is to load your finished filter stacks (LRGB, or SHO, or all 7!) into the Blink tab. There, tick off the Align Images checkbox. And then over on the far side use the Export Aligned button. The result is the same, you will get new files with "_aligned" at the end of the name. And those are now good for ST compose, though again with a possible caveat about auto-reading the total exposure time correctly.
Method 2 might be more intuitive. Whichever works for ya.
Anyway, it's not necessarily super optimal per Ivo, but since ASTAP does not allow manual override of reference selection, post-stacking registration is the only option I'm aware of (absent, perhaps, running cal and align only on all the subs first, and then stacking those - but I'm not 100% sure on that in the multiple mono filter scenario).
Re: ASTAP and Startools - Mono Imaging Help !
Thanks Mike ,
Got it !!
Your a fair dinkum champion, mate ( Aussie slang for great guy )
Your explanation provides a clear understanding now of the basics in ASTAP.
Next job is to write up a brief procedure, as I do with all my Astro related tasks. You never know when you need to refer to notes or help other folk , just like you have
Clear Skies
Martin
Got it !!
Your a fair dinkum champion, mate ( Aussie slang for great guy )
Your explanation provides a clear understanding now of the basics in ASTAP.
Next job is to write up a brief procedure, as I do with all my Astro related tasks. You never know when you need to refer to notes or help other folk , just like you have
Clear Skies
Martin
Re: ASTAP and Startools - Mono Imaging Help !
Hi Mike,
A happy Astrophotography now
Have another question for you ……..
Say I imaging an emission nebula and captured 4hrs of Ha under 20% Moon ( night 1) and then a week later capture 4hrs of Ha under 90% moon ( night 2 )
How would you stack this in ASTAP considering vastly different sky brightness ??
Would you stack night 1 using Sigma , then stack night 2 using Sigma. Then combine both and stack again using Average ??
Your thoughts and expertise
Martin
A happy Astrophotography now
Have another question for you ……..
Say I imaging an emission nebula and captured 4hrs of Ha under 20% Moon ( night 1) and then a week later capture 4hrs of Ha under 90% moon ( night 2 )
How would you stack this in ASTAP considering vastly different sky brightness ??
Would you stack night 1 using Sigma , then stack night 2 using Sigma. Then combine both and stack again using Average ??
Your thoughts and expertise
Martin
-
Mike in Rancho
- Posts: 1166
- Joined: Sun Jun 20, 2021 10:05 pm
- Location: Alta Loma, CA
Re: ASTAP and Startools - Mono Imaging Help !
Hi Martin,
How did you handle similar situations with DSS and OSC/duoband?
But, hopefully things like Ha and SII have enough blocking that by and large the moon would not be an issue. OIII of course is very susceptible to moon.
But either way, I would just throw everything into the pot. If the sessions have their own flats, you would probably be choosing classify by date for master creation. Assuming identical capture settings on the two nights, your appropriate bias and darks or master dark should be fine.
Unlike with DSS, in ASTAP there is no way to turn off global brightness normalization during stacking. But to me that's probably a good thing - when I was using DSS I ultimately ended up turning that back on even though it was against recommendations. And sigma rejection, other than those pixels ejected for being outside the set SD, thereafter averages anyway, so really no need to do the same thing in two different steps, plus you would lose global sigma rejection per a single reference.
You could experiment of course with such extremes of moon or no moon and see how things turn out. Stack each night separately, then one all together, and see how they compare. I would surmise that the two-night stack with greater integration will be superior.
Things like this are where more bells and whistles in the stacker can start to possibly be helpful. Both WBPP and I believe Siril will allow for subframe weighting based on quality. And WBPP of course has local normalization. In a thread earlier this year Ivo gave his tips for when using LN would be useful, and I believe different session conditions was one of those.
You can think of LN as sort of an SVD but for the brightness normalization that is requisite to sigma rejection stacking. A best reference is created and then all the subs are locally (spatially) normalized to that. This can aid in reducing complex gradients that can occur if capturing during different brightness nights, or even a long imaging sweep by the mount which shifts the LP regions, and thus - in theory - makes things better for Wipe.
How did you handle similar situations with DSS and OSC/duoband?
But, hopefully things like Ha and SII have enough blocking that by and large the moon would not be an issue. OIII of course is very susceptible to moon.
But either way, I would just throw everything into the pot. If the sessions have their own flats, you would probably be choosing classify by date for master creation. Assuming identical capture settings on the two nights, your appropriate bias and darks or master dark should be fine.
Unlike with DSS, in ASTAP there is no way to turn off global brightness normalization during stacking. But to me that's probably a good thing - when I was using DSS I ultimately ended up turning that back on even though it was against recommendations. And sigma rejection, other than those pixels ejected for being outside the set SD, thereafter averages anyway, so really no need to do the same thing in two different steps, plus you would lose global sigma rejection per a single reference.
You could experiment of course with such extremes of moon or no moon and see how things turn out. Stack each night separately, then one all together, and see how they compare. I would surmise that the two-night stack with greater integration will be superior.
Things like this are where more bells and whistles in the stacker can start to possibly be helpful. Both WBPP and I believe Siril will allow for subframe weighting based on quality. And WBPP of course has local normalization. In a thread earlier this year Ivo gave his tips for when using LN would be useful, and I believe different session conditions was one of those.
You can think of LN as sort of an SVD but for the brightness normalization that is requisite to sigma rejection stacking. A best reference is created and then all the subs are locally (spatially) normalized to that. This can aid in reducing complex gradients that can occur if capturing during different brightness nights, or even a long imaging sweep by the mount which shifts the LP regions, and thus - in theory - makes things better for Wipe.
Re: ASTAP and Startools - Mono Imaging Help !
Mike,
Thanks
In DSS I just followed the ST prerequisites, No background calibration, however I was dealing with vastly higher noise gradients using the 2600MC and Dualband filter than this new 2600MM and bonafide NB 3nm filters
So far ASTAP is working well and looking forward to stacking more data sets
It looks like a more professionally designed suite than DSS which had a big upgrade recently but it’s still the same stacker at the end of the day.
As far as stacking multiple nights data from same filter in ASTAP
I assume you would stack each night ( incl calibration frames ) as per previously documented, and then combine them again in stack mode but select “alignment and calibration”
Then stack them again using Sigma ??
Not sure if this would work ?
Also I used ASTAP to align my Ha and Oiii stacks for loading into Compose ST ( Sii was a noise bomb so ditched it ) Helix nebula has very little Sii signal
1st Option worked well ( Calibration and Alignment)
2nd Option didn’t work at all ( Blink , Align images and Export ) Nothing happened ?
I noticed I’m using an ASTAP version in 2021 , maybe it’s a good idea to download the latest version ( Win 10 64 bit )
Martin
Thanks
In DSS I just followed the ST prerequisites, No background calibration, however I was dealing with vastly higher noise gradients using the 2600MC and Dualband filter than this new 2600MM and bonafide NB 3nm filters
So far ASTAP is working well and looking forward to stacking more data sets
It looks like a more professionally designed suite than DSS which had a big upgrade recently but it’s still the same stacker at the end of the day.
As far as stacking multiple nights data from same filter in ASTAP
I assume you would stack each night ( incl calibration frames ) as per previously documented, and then combine them again in stack mode but select “alignment and calibration”
Then stack them again using Sigma ??
Not sure if this would work ?
Also I used ASTAP to align my Ha and Oiii stacks for loading into Compose ST ( Sii was a noise bomb so ditched it ) Helix nebula has very little Sii signal
1st Option worked well ( Calibration and Alignment)
2nd Option didn’t work at all ( Blink , Align images and Export ) Nothing happened ?
I noticed I’m using an ASTAP version in 2021 , maybe it’s a good idea to download the latest version ( Win 10 64 bit )
Martin