Am I getting enough data in my images?

[Spleenus]

Hi all! I've been diligently trying to read as much as I can, and watch some youtube videos to try and solve my processing of images 'problem' with photoshop.

In a nutshell, when I first start doing the stretching of the curves I just cannot get some of the data to budge far from the left hand side of the histogram. I end up distorting what data I do have as something chronic and lots of gradients appear in the image. From then on I'm just fighting a losing battle. This battle doesn't happen when I use some of the 'training images' which you can use whilst watching the video. Like Astrobackyards or astrofarsography's.

I realise I am starting off at a difficult level, as I'm trying to image some galaxies (M83) with a 150mm Maksutov and QHY163c. The subs were 6 minutes long. So I am suspecting that I'm simply not getting enough data through (long focal length?) to the camera. And thus not having enough data to stretch.

Am I on the right train of thought? Thanks for any help or tips which might be pointed in my general direction :)

[Zuts]

posted by mistake

[Zuts]

Hi,

It may be helpful if you post the image so people can look at it.

Also, depending on your optical train and light pollution there could be lots of dust bunnies and gradients. If these are not removed by taking appropriate flats then when you stretch these will become apparent and dominate the image.

It is also useful to take darks and bias frames and give them to whatever stacking program you are using, before you begin your PS processing. Are you taking flats, darks and biases.

Also at 1800 mm focal length your guiding and polar alignment needs to be very very accurate. If it's not the data will not be very good. How round are the stars?

Finally, F13 is a very slow scope. You will need far more data than 6 minutes I would think. I get a great image after 3 minutes at F7 120 mm refractor. You have nearly twice my focal length so the general rule is double the F-Ratio, 4x the exposure time so are looking at 12 minutes to gather the same amount of light as me.

You are potentially being very optimistic starting off at F13 as a beginner. Most beginners start of with something like an ED80 and a focal reducer to learn the basics and then pick an image scale that they like and continue on.

Cheers
Paul

[jahnpahwa]

Crank up the gain on the camera. :thumbsup: I don't really see why high gain can't partially offset a slower scope.

But also, post an image and maybe a screen shot of your histogram before and after a stretch? And maybe hit Centaurus a instead of M83 while dialling in systems. M83 is relative bright as far as galaxies go, but is still bloody faint, even in an f5 scope it takes a lot of integration to see much. You'll see what's happening and what needs fixing quicker with cent A if it has to be a galaxy, but otherwise Carina neb will throw up results quickly and you'll troubleshoot more quickly.

[AdamJL]

Hi Darren
What program are you using? I found something similar when using Photoshop, but using a 32bit file. Doing a Levels stretch didn't move the histogram much. When I used Photoshop, I would do a stretch in 32 bit a few times, but then move onto 16 bit for more stretching and the majority of the work.
That said, it would be good for you to share some data so we can take a look!

edit: the above points are good as well. Increase your gain, but also your subs times as you're shooting at f/13.
That said, you should still have some useable data even at f/13 with 6min subs if you've got enough signal. What was the total integration time?

[glend]

Your scope is far too slow for imaging. There are plenty of cheap fast imaging newts around at f4 or f5. You will make much better use of your time with a faster scope, and it will accelerate your learning the craft.

[Spleenus]

Some good feedback there for me to consider so thank you all :)

I have done bias and darks, I didn't do flats but I can see the effects of that in the image on stretching. I was comfortable with missing them for now (they're a project for another day :) ). I had about 2 hours worth of integration time and was adjusting in 16bit.

I've also spent a fair bit of effort on guiding and the polar alignment. So the stars are round, it's in focus and I'm quite happy with it for now. I guide off-axis because of the focal length.

I probably didn't mention that I am clipping the blacks and losing data which is probably half my battle. I also, now I've since done research, realise that the gain I had is quite low compared to what other people have been using for this camera. I had it set at 80, but I've seen lots of people using 200 to 400 gain. So that's definitely another area to look at when the clouds go away for another night as it should definitely give more data to play with.

I also do have a 10" reflector, and the brackets to mount it on the mount. My primary interest is not necessarily getting the wide field views, but in planetary nebula and galaxies (I am a physics, astrophysics graduate) so I am well aware of the challenge I have presented. But I have the tenacity to keep going. Plus, the planets aren't up in the mean time for me to play around with. Hence why I got this scope with the focal length (and in my budget :) )

I'll post the image I did have shortly. I was able to make out the dust clouds in M83, but maybe I am expecting too much.

[Spleenus]

Ok, this is what I've ended up with. The image is just noisy as I've had to super stretch it. I do see the result of no flats in the image. I was thinking I just didn't have enough data as the histogram you can see is what it is like after doing the stretching and some levels adjusting. I've not done any other processing. That just makes it go more horrendous (if it's possible! :) )

And, on the off chance anyone wants to see the initial stacked image here is the link
Onedrive image link (https://1drv.ms/u/s!AmdNzEQMlqiOgdBBvNqSKVtKHkTLUg?e= VxuPak)

[Zuts]

Hi Darren,

You have some data there but i think using some flats would help. Your polar alignment and guiding look good. I have done a quick process of the stack in PixInsight, except for colour balance I don't think my version is any better than yours, but I can't seem to get rid of the banding.

[multiweb]

I'm also having so much trouble getting good data from my backyard on this one that I have come to the simple conclusion that everybody has been imaging it and there's not a single photon left in it.

[Zuts]

Yeah, they are not dust bunnies, they are actual holes where all the photons are missing!

[Nikolas]

You need to collect more data 6 minute subs are fine but lots more of them by the look of it it is a little noisy, did you use a light pollution filter? Also is the image you provided a crop or did you bin it?

[ChrisV]

I'm with Glen on this. I reckon you are making it hard for yourself. Imaging DSOs with a Maksutov is really hard. It's very slow so you will need really long integration times too get anything. You want to be f6 or preferably less. My refactor is f5.9 and my newt F5. My first dso was M83 with the F5, 4 hours total and now I think that's a bare minimum.

You haven't said anything about your data acquisition. How many 6 min subs. Darks, flats? Processing?