Specify reference frame?

Questions, support and discussions
szymon
Posts: 28
Joined: 29 Aug 2019, 18:17

Specify reference frame?

Post by szymon » 30 Aug 2019, 00:28

Hi there,

I absolutely love ASTAP for stacking, thank you for a great piece of software. I was pointed here by the folks on the NINA discord, when I was having trouble with Deep Sky Stacker, and after playing with it for a while I think that ASTAP is simply superb.

I was wondering if you can explain how I can tell the software to use a specific frame as the reference image against which to align all of the other frames?

My use case is that I want to produce four separate stacks -- Luminance, Ha, Oiii, Sii -- but have them all aligned the same, so that when I combine them (e.g. in StarTools) they all line up together perfectly. I can't see a way of doing this in ASTAP; currently it chooses which frame to use as the reference image by selecting the one with the best HFD in each stack.

Is what I want to do possible? If not, could it be added as a feature?

Many thanks,

-simon

han.k
Posts: 289
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Specify reference frame?

Post by han.k » 30 Aug 2019, 08:40

Hello Simon,

ASTAP uses the images with the lowest HFD value as reference. So for combination of red, green and blue it first combines red (against the red frame with the lowest HFD), green (against the green frame with the lowest HFD), blue (against the blue frame with the lowest HFD). Then it combines the resulting red, green and blue stack. Same for r,g,b,l or Luminance, Ha, Oiii, Sii.

At the moment, I don't see a reason for manually setting a reference frame but please explain to me why this would be beneficial.

If you have an image series where I could experiment with, that always helps.

Han

szymon
Posts: 28
Joined: 29 Aug 2019, 18:17

Re: Specify reference frame?

Post by szymon » 30 Aug 2019, 09:34

Hi there Han,

The reason to manually set a reference frame is so that I can get separate red, green, blue stacks and combine them in different software. I don't want to merge them in ASTAP, I want to merge them in StarTools (so that I can control and fine tune how they are merged as part of the development process). I just want all of the frames using exactly the same reference frame. This is a pretty common use case for other stackers.

I will upload some data for you in a moment.

-simon

szymon
Posts: 28
Joined: 29 Aug 2019, 18:17

Re: Specify reference frame?

Post by szymon » 30 Aug 2019, 09:52

Here is my data for the elephant's trunk nebula:

https://www.dropbox.com/s/bjmcvax36ele6 ... a.zip?dl=1

There isn't much data here, this is more of a test image than anything else. There are light frames with four different filters:

12 x 300s CLS (Luminance)
18 x 300s Ha
6 x 300s Oiii
6 x 300s Sii

I also took darks, flats and darkflats -- I've uploaded the master dark and master flat created by ASTAP.

Each set of lights should be stacked into its own stack. More importantly, all four stacks should be aligned to each other, so that I can combine them in other tools.

Does that make sense now?

-simon

han.k
Posts: 289
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Specify reference frame?

Post by han.k » 30 Aug 2019, 12:01

ASTAP produces autmatically stacks for each colour. You can see them in the results tab. I will have a look to your data later this day. First I have to finish some other changes.

Clear skies, Han

szymon
Posts: 28
Joined: 29 Aug 2019, 18:17

Re: Specify reference frame?

Post by szymon » 30 Aug 2019, 12:08

I know it produces the stacks -- and they're great :-). But the stacks are not aligned with each other :-\

When ASTAP creates the final master stack to create its combined image, it has to run its algorithms to align the stacks:

Code: Select all

19:25:25  Reference image, largest with best HFD is: E:\TARGETS\TRUNK\LIGHT\2019-08-27_03-59-10_CLS_19.70_300.00s_1@  20xFD  20xF  4xD  12xCLS_stacked.fit
19:25:27  Adding red file: "E:\TARGETS\TRUNK\LIGHT\2019-08-27_03-51-44_Sii_19.60_300.00s_6@  20xFD  20xF  4xD  6xSii_stacked.fit"  to final image.
19:25:30  15 of 16 tetrahedrons selected matching within 0.005 tolerance.  Solution x:=0.999367*x+ -0.000452*y+ -2.075692,  y:=0.000509*x+ 0.999167*y+ 15.561219
19:25:31  Adding green file: "E:\TARGETS\TRUNK\LIGHT\2019-08-27_03-16-37_Ha_19.70_300.00s_12@  20xFD  20xF  4xD  18xHa_stacked.fit"  to final image.
19:25:34  20 of 23 tetrahedrons selected matching within 0.005 tolerance.  Solution x:=0.999471*x+ -0.000823*y+ -3.060699,  y:=0.000753*x+ 0.999395*y+ 27.953323
19:25:35  Adding blue file: "E:\TARGETS\TRUNK\LIGHT\2019-08-27_01-58-58_Oiii_20.50_300.00s_6@  20xFD  20xF  4xD  6xOiii_stacked.fit"  to final image.
19:25:38  11 of 13 tetrahedrons selected matching within 0.005 tolerance.  Solution x:=0.999934*x+ -0.000586*y+ -5.993951,  y:=0.000460*x+ 0.999645*y+ 31.839029
My other software isn't that clever! That's why I want the stacks already aligned to the same image, so that the other software can create the combined image. (While the other software isn't clever enough to automatically align the stacks, it is very good at composing an image from layers, giving full control over things like the intensity of each channel, interpolating channels which aren't there, synthesising luminance frames from RGB, etc).

I completely understand you have other priorities :-)

han.k
Posts: 289
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Specify reference frame?

Post by han.k » 30 Aug 2019, 14:20

szymon wrote:
30 Aug 2019, 09:52
12 x 300s CLS (Luminance)
18 x 300s Ha
6 x 300s Oiii
6 x 300s Sii
Okay I understand your point. If you combine in ASTAP the three filters (CLS, Ha, Oiii) as RGB, then the three stacks are aligned and in the three R,G,B layers. You could split them or use the directly.

ASTAP can't process four colour channels. You could process CLS,Ha,Oiii as RGB and CLS, Sii as RG. Then you would have 2 result files with totally 4 channels aligned. If I remember well it will align on the red channel

I have to think about this a little longer. Maybe it is possible to adapt it to four channel processing. But your software, can it process a four channel FITS or do you need separate files?

I have look to your files first.

Han

szymon
Posts: 28
Joined: 29 Aug 2019, 18:17

Re: Specify reference frame?

Post by szymon » 30 Aug 2019, 14:40

My software StarTools has a 'Compose' module which requires four separate files to fit in the channels. That said, the developer is very responsive (he replies on https://forum.startools.org/ just as fast as you do here!), so he would probably add support for four-channels in a single FITS file if I asked. You could try out the software for processing if you wanted, the demo at https://www.startools.org/downloads is completely free and not time limited ;-)

If we could just specify a specific file for ASTAP to use as the reference (or if when processing multiple filters it could choose the one file with the best HFD from all filters and use that as the reference for all filters) then that would be a great solution :-). For now however your workarounds sound sensible, I will try those and then split the combined files out into separate files and see if I can get it working like that!

Thanks,

-simon

han.k
Posts: 289
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Specify reference frame?

Post by han.k » 30 Aug 2019, 15:13

I will experiment with your files and see if I see an other solution.

han.k
Posts: 289
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Specify reference frame?

Post by han.k » 30 Aug 2019, 16:33

The signal in your images is pretty weak. I looking currently to this webpage for so sample material e.g. IC1396a:

http://chandra.harvard.edu/photo/openFI ... _data.html

All images are in a separate FITS files but aligned. Multi layers FITS doesn't seem to be much used in astronomy.

In principle the blink tab can already blink your four filter channels aligned. It aligns to the first image. Maybe export to aligned images would be something.....

Post Reply