Help with Noise in M31 images

[nucdoc]

Below is preliminary data on M31.

Taken with a Star 71, Astronomik CCD LRGB filters, Astronomik 6nm Ha filter and Atik 383l+

I am took the Ha under a full moon, the LRGB were taken in Moon free skies, 1.5 hrs of L, and 1 hour of Ha, RGB.

Ha 20 minute subs LRGB all 10 minute subs

Flats, darks bias all used in PI, lots of DBE, etc...

 

Part of the background noise is likely related to the low position of M31  during the time I shot the images, It is not high overhead till midnight at my latitude and I shot these at 9:30 to 11:30 pm.  Could I expect to get better signal to noise with a couple more hours per channel, or should I also try to image only when M31 is high overhead, in the darker part of the sky

 

THanks

Mark

 

[Xplode]

If you have lots of light pollution you should definitely wait till it's high in the sky, if you don't have much light pollution it doesn't matter that much if it's very high.

A lot more Luminance data would make the most difference, but i suggest you also get at least another hour of RGB data.

To help with noise i suggest you start dithering if you aren't already doing dithering.

[Adam J]

From the look of it you have a colour balance issue, too much red and too little blue. That dark ring around M31 should be showing up blue. 

you say: 

"1.5 hrs of L

1 hour of Ha, RGB.

Ha 20 minute subs LRGB all 10 minute subs"

So does that mean 1 hour each of Ha, R, G and B? or 1 Hour total of Ha R G and B?

I am guessing that the Red + Ha are really dominating the colour balance, hence the red tint at your black point, this is causing the blue areas not to show up as its being dominated. 

Draw down the red until you get a black background and then re-stretch. Better balance will also most likely help with the noise in the red channel. 

[wimvb]

The best way to reduce noise is of course always to have more data. With the data you got you can try the following

1. Make a master luminance from just the L frames.

2. Make a master luminance from ALL light masters: L, R, G, B, and maybe even Ha

Evaluate both luminance masters and use the one with best SNR.

After (Ha)RGB combination, do an extreme noise reduction on this. Or simply blur the image.

Then combine your best L master with the blurred/noise free RGB image

This is a variation of the technique described in Warren Kellers book (inside PixInsight)

 

Good luck

[nucdoc]

Thanks guys, for all the helpful suggestions...

Xplode,

Yes I have a lot of light pollution, especially to the NE where there are street lights, so I guess waiting for it to be higher will be a good idea.  I have an LX850 mount, no way to dither using the star lock guide, but I could switch back to my guide scope if I needed to.

Adam3,

I have 1.5 hours of Luminance, and 1 hour of each of the other channels, RGB, Ha, but I need more and will collect it after the skies clear, hopefully before the next lunar cycle

 

WimVB,

I followed your advice, using the Master luminance from the L frame, when I used the RGB and RGBha the SNR was not as good.

I then did aggressive NR on the (Ha)RGB and added this to the L, followed by DBE, Masked stretch, Curves, Saturation, below is the result, much less noise in my opinion

Thanks again

Mark