Which is worse, my data or my processing?

Questions and answers about processing in StarTools and how to accomplish certain tasks.
Post Reply
Skipper71
Posts: 3
Joined: Fri Feb 12, 2021 4:41 am

Which is worse, my data or my processing?

Post by Skipper71 »

The easy answer is both and I'm in no position to disagree, but some advice from someone who has actually done this for more than a month would be appreciated.

I'm VERY new to this and attached is the total of what I have produced as an astrophotographer. I'm using Astro Pixel Processor and StarTools for post processing.
M45-NoSt4.jpg
M45-NoSt4.jpg (285.22 KiB) Viewed 3424 times

1. It's very exciting. 2. I know it needs vast improvement 3. I don't know what to do/how to improve it. I SEEMS to me I had to stretch it a lot to get the color out of it. It SEEMS to me I have a lot of noise in it, the "smudges" in the background are especially unappealing. I would be very grateful for a critique and some advice on what to target as priorities for improvement, be they in acquisition or processing.

Askar FMA180, ASI533MC Pro, 100 gain. Baader UV/IR -Cut/L. (I know this setup is significantly undersampled. Would using drizzle in APP be a good idea? If so, ... double resolution then bring it back down with binning in StarTools? I'm not sure at all about choices to be made if using drizzle.)

166x 60s (dithered) lights. Processed top 151 in APP using darks, flats, dark flats and BPM. I followed all StarTools guidance for APP.

Link to fits here: https://1drv.ms/u/s!AnPGv8pl3Icnow0mPuM ... R?e=6eeodd

StarTools workflow -
Loaded as Linear, from OSC/DSLR...
AutoDev
Bin at 50%
Crop
Wipe: Gradient Aggressiveness to 85%, Vignette
Film Develop (because autodev doesn't give me enough color) to 97%, Gamma to 1.20.

Question: Do I simply not have enough data in my stack?

Contrast - default
HDR - Reveal Core
Sharp - auto generate mask, default
Color - Saturation amount 500%, Dark Saturation 1.0, RGB Bias reductions at 1.39, 1.79, 1.10 respectively.
Shrink - manual mask: auto, stars, source "Stretched", Filter Sensitivity 30, do. Keep, Dim.
Super Structure - Isolate
Noise Reduction - default

Did I make any poor choices in my workflow? could I have made any better choices?

Many thanks to anyone willing to help!
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: Which is worse, my data or my processing?

Post by almcl »

Hi Skipper

Welcome to StarTools. You'll get very good advice from Ivo when he happens by, but in the meantime here are some very non-expert comments. Bear in mind that what is really important is whether you like your images, so my opinions don't really matter very much.

That said, on my monitor the image looks a little over saturated - that's partly a personal taste thing - but one thing I do notice is that there are quite a few green stars. You're probably aware of the phrase 'there are no green stars in space' but here's the bottom right of your image blown up a bit but otherwise untouched:
Untitled-1.jpg
Untitled-1.jpg (25.81 KiB) Viewed 3418 times
I am not sure if you used it or not, but in the colour module the Cap green slider might help alleviate this. Having said that I had a quick go at your data, which looks pretty good, and didn't get quite much green:

M45.jpg
M45.jpg (168.34 KiB) Viewed 3418 times
You may not care for this rendition, but see what you think?
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
Skipper71
Posts: 3
Joined: Fri Feb 12, 2021 4:41 am

Re: Which is worse, my data or my processing?

Post by Skipper71 »

That's true about my opinion is what is REALLY important and why I'm looking for a little coaching. I don't like my picture very much. I recognize it's my first attempt so I'm happy I got what I got, but can see plenty of room for improvement. The color is there but I felt like I was over-saturating it but couldn't find any other way to bring that color out. Your background is MUCH better, more even, than mine. Is that because I over-saturated it to get the blue? (and did I have to do that because my data was insufficient? Or is there another way to get there without ruining the background... IF that was the cause? Ahhh, questions!)

That's a great point about the green stars, I did not (obviously) use the Cap green slider. I'm taking notes...

Thank you very much for taking the time!
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: Which is worse, my data or my processing?

Post by almcl »

Noise in the background is a perennial for all AP, it seems.

I find it best not to try and mitigate it all in one go. First thing is to increase the Dark Anomaly Filter in Wipe to maybe 4 or 5 px (works for me but your data may need more or less) and then use the 2nd Autodev with an RoI (worth adjusting this to see what difference it makes) and then adjust the 'Ignore Fine Detail' and 'Outside RoI Influence' sliders, balancing a cleaner background with nebulosity strength. Later on, use 'isolate' in the Life module (possibly with a mask) and maybe also 'Dim Small'. Finally when turning tracking off worth playing with the Grain size.

If the result is still not right, it is possible to use the layer module to remove stars, then work on the background before adding the stars back in (see here for a detailed description viewtopic.php?f=7&t=244. )
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
Burly
Posts: 387
Joined: Mon Aug 31, 2015 7:06 am
Location: Northamptonshie uk

Re: Which is worse, my data or my processing?

Post by Burly »

Data seems fine , ,no cap green not required as you adjust green channel to suit ,colour balance using stars via mask used ,wipe i knocked down slightly to avoid wiping nebulosity ,and upped gamma to 1.3 to illuminate nebulosity ,use the RGB button in colour module so you can see the colour channels and that will give you an idea where the green bias is,more data the better .
Attachments
M45-NoSt_Easy-Resize.com.jpg
M45-NoSt_Easy-Resize.com.jpg (96.26 KiB) Viewed 3374 times
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Which is worse, my data or my processing?

Post by admin »

Hi,

Some great advice and observations here already.

That said, this caught my eye;
I had to stretch it a lot to get the color out of it.
and
Film Develop (because autodev doesn't give me enough color) to 97%, Gamma to 1.20.
That's not how StarTools works; luminance (detail) and chrominance (color) is processed separately in StarTools. Your stretching has no bearing at all on the coloring (which is also why there is no need in ST for hacks and sub-optimal stretches like ArcSinH etc. to desperately try to preserve coloring). Use the Color module towards the end of your processing, as shown in most workflows and as outlined by the module buttons on the home screen when read from left-to-right, top-to-bottom.

Definitely consult the Color module documentation, which shows you various ways to color balance your image.

What is throwing off the default color balance at the moment, is the rather sever chromatic aberration in your optical train. It is causing the tell-tale blue/purple stars. This is somewhat puzzling/concerning, as the Askar FMA180 is supposed to be apochromatic and should not show meaningful chromatic aberration in an image like this. Yet here we are... :cry: Perhaps there is some other piece of low quality glass that is causing the chromatic aberration in your optical train? If not, I would contact the manufacturer, as I would not deem this acceptable of a scope advertised as apochromatic and boasting "excellent correction of chromatic aberration".

In situations like these, use the MaxRGB technique in the documentation to achieve a good color balance in the nebulosity. The nebulosity is blue, extending to gray and brown. Green dominance should be balanced out, which you can achieve as well by simply clicking on any areas that appear green dominant (some small bits of green here and there are OK).

Do not use Cap Green unless you are absolutely sure your color balance is otherwise spot-on, as it is not a color balancing tool; it is a last resort.

You can use the Highlight Repair setting to mitigate the worst of the purple in the core of the stars. The Filter module's Fringe Killer mode can be used in conjunction with a mask to remove any further purple coloring around the stars.

Your dataset does not appear undersampled to me; non-overexposing stars show multi-pixel stellar profiles, and deconvolution indeed tightens them and brings out (some) more detail.

Hope this helps!
Ivo Jager
StarTools creator and astronomy enthusiast
Skipper71
Posts: 3
Joined: Fri Feb 12, 2021 4:41 am

Re: Which is worse, my data or my processing?

Post by Skipper71 »

Many thanks! That's the kind of constructive coaching I needed. :thumbsup:
ionia23
Posts: 62
Joined: Fri Jul 16, 2021 7:39 pm

Re: Which is worse, my data or my processing?

Post by ionia23 »

I look at this and I think "He's been doing this a month, and it already looks more impressive than things I've seen by people who've been doing it for years".

Wow.
Post Reply