Help with processing M33

[Arrowstar]

Hi everyone,

Last night I managed to get some time on M33, and as a beginner to the hobby who is recently coming back to it after taking the (very busy!) summer off, I could use some help.

I was able to capture ~1 hour of data on M33, and while I intend to grab more, I wanted to see what I could do with what I have. Turns out, the answer is not much. I'm sure some of the problem is that I simply don't have enough integration time, and I'm looking to rectify that on the next clear night. In the mean time, I was hoping that perhaps someone from here might be able to take a look and either help me understand how to process it better or tell me to come back when I've got a workable amount of data!

Some details of the stacked image:

1) Image stacked in DSS
2) Total integration time 59 minutes
3) Sub-exposure length 1 minute
4) ISO 400 (had to reduce from ISO 800 to get the data back down to 1/3 histogram)
5) Shooting from a red zone (ah, light pollution!)

Here is a link to the stacked image straight from DSS (140 MB): https://dl.dropboxusercontent.com/u/291 ... _59min.TIF

What do you think? Can I do something with that I have or do I just need to gather more integration time before processing (well, I need to do that anyway! )?

[almcl]

M33 is quite a spread-out target and quite faint as a result. 60 secs at ISO 400 is going to be a challenge to get the signal out of the noise.

Can you get to a less light polluted site for imaging, or if not and your light pollution is mainly caused by Sodium street lamps, consider a light pollution filter? This does remove some spectral lines but greatly increases the capture time that is possible.

[Arrowstar]

M33 is quite a spread-out target and quite faint as a result. 60 secs at ISO 400 is going to be a challenge to get the signal out of the noise.

Can you get to a less light polluted site for imaging, or if not and your light pollution is mainly caused by Sodium street lamps, consider a light pollution filter? This does remove some spectral lines but greatly increases the capture time that is possible.

Yes, I do. I did not think to put it in until I was most of the way done with my imaging session, but I will for next time. What do you believe the minimum exposure duration is to be able to successfully capture M33?

[almcl]

Bear in mind I don't know what scope, camera or mount you are using, but when I imaged it this time last year, I used 180 second subs at ISO1600. This was with the Astronomik CLS-CCD clip filter on a Canon 700d attached to my 8" Skywatcher. These produced a histogram that was barely 25%. A 300 second one was about 40%.

Here's a reduced but unstretched jpeg of one of the 180 sec subs. As you can see, M33 is still fairly faint.

6195.jpg (80.38 KiB) Viewed 9690 times

[Arrowstar]

Bear in mind I don't know what scope, camera or mount you are using, but when I imaged it this time last year, I used 180 second subs at ISO1600. This was with the Astronomik CLS-CCD clip filter on a Canon 700d attached to my 8" Skywatcher. These produced a histogram that was barely 25%. A 300 second one was about 40%.

Thanks for the info. My scope is an AT72ED and I image with a Canon T3. I do have the CLS-CCD clip filter you mentioned and have it installed in the camera now. Last night was actually my first ever successful attempt at auto-guiding (QHY5L-II on an Orion guidescope). I was able to demonstrate two minute subs without trails, but I will look into pushing 180 seconds. Mind you, I'm shooting from really bright skies (red zone with a white zone city near by), so I may not be able to do the full three minutes. We'll see. I guess we'll find out if that LP filter lives up to its name!

By the way, is lowering ISO to achieve a longer exposure time a valid solution to keeping your histogram at 1/3? As I understand you gain dynamic range at the lower ISOs but at the expense of noise. What are your thoughts?

[almcl]

I am no expert but once you lower the ISO below the unity gain point (which is different for different cameras) in the words of Steve Richards 'you are wasting photons'.

Having said that, every set and circumstances are different and what works for one doesn't always work for another. But if sky glow is washing out the signal, you have to do something, I can barely get 120 seconds without the clip filter on some nights. But I think pushing the histogram to 40 or 50% might be worth a shot?

[Guy]

Hi

I'm no expert either but for what its worth I agree.

In my (limited) experience:
- Although ideally I avoid saturating pixels because it is lost data - in reality I often saturate some parts of the image. It is usually a compromise. Most often it doesn't cause a major problem - so work on capturing the faint signals and, when you have that sorted, see what sort of a problem you have at the high end.
- You can't make a decent image if you don't capture the signal photons in the subframe - no matter how many subs-frames you take. So, to my mind, you need to think first about the signal, then the noise - so even in difficult light conditions (e.g. light pollution, moonlight) make sure you have exposed enough to capture sufficient signal so that stacking sub-frames can help increase the SNR.
- I'd experiment with a range of exposures (as well as lower ISO settings - as you have started to do) - from highly over-exposed (e.g. 5 mins) downwards - to find what the optimum setting is.
- This external article describes using this 'Expose to the Right' approach - in this case when photographing the milky way - which shows the technique and what can be done even under difficult light pollution conditions.
- Image processing software like StarTools can do a lot to get rid of noise (e.g. the Wipe module) - as long as there is a reasonable amount of signal to work with.
- Using a LP filter may help - but you may want to read Ivo's StarTools article about colour balancing when using a LP filter.

I hope this helps - I'll be interested to see how you get on.

Guy

[Arrowstar]

Hi everyone,

I've managed to get a few more hours in on M33 and have taken a look at the results. Unfortunately my data is still pretty grainy and I'm wondering if I can't get some help with it.

Here is a link to the stacked image: https://dl.dropboxusercontent.com/u/291 ... ntropy.TIF

(This image uses the entropy method for stacking. I will post another image that uses the Kappa-Sigma method in a moment. EDIT: Here it is: https://dl.dropboxusercontent.com/u/291 ... aSigma.TIF)

Some details:

1) 2 hours of 3 minute subs @ ISO400 + 1 hour of 1 minute subs at @ISO 400
2) All light frames calibrated with 50 darks, 50 flats, and 100 bias frames.
3) Telescope is an AT72ED with a Canon T3 DSLR camera. I used a clip light pollution filter on the 3 minute sub data.
4) Shooting from a red zone.

What do you think? Is what I have savable or do I need to get more subs?

[Guy]

Hi,

Can you try just stacking the 3 minute subs? It may be the 1 minute subs are degrading the image.
Can you also post one of the 3 minute subs as well? I'd like to see what has been captured.

Thanks,

Guy

[Arrowstar]

Hi,

Can you try just stacking the 3 minute subs? It may be the 1 minute subs are degrading the image.
Can you also post one of the 3 minute subs as well? I'd like to see what has been captured.

Thanks,

Guy

Sure, I would be happy to.

Here is a stack of just the 3 minute subs: https://dl.dropboxusercontent.com/u/291 ... acked6.TIF

Here is one 3 minute sub-exposure: https://dl.dropboxusercontent.com/u/291 ... n_0050.CR2

Thanks for the help!