Strong grain in stacked light pollution?

[matd]

Hello to all! I'm pretty new to astrophotography. I'm imaging with an astro-modified Sony A5000 and an unmodified NEX-5, and a bunch of old M42 manual telephoto lenses. I have a Star Adventurer camera mount.

I sometimes seem to be getting a strange grain in the light-polluted background of my images, which is not present in the individual exposures. The grainy effect makes it hard to remove in postprocessing. I've accentuated it a little here. This is the result of stacking around 200 30-second subs in DeepSkyStacker:

I'm thinking it's either high-level cirrus, or a noise feature of the camera. It only appears after stacking, and gets more accentuated with more stacked images. The direction of the grain varies in different images, which makes me think it might be an effect of stacking images with faint clouds?

All and any advice would be much appreciated!

Kind regards

-Mat

 

 

 

[D4N]

Are you using kappa sigma clipping in DSS?

Are you also using calibration frames?

At a guess I would say that DSS has stacked the noise and it looks odd because the image shifts between frames but it certainly does look strange.

 

Otherwise double check all the subs frames, especially the last ones to check there was no dewing or other problems that are getting thrown in.

[sharkmelley]

It is so called walking noise.  The stars drifted across the frame during the course of your imaging session so when you align the stars in those frames it appears that the hot/warm pixels are walking across the frame instead.  A combination of hot pixel detection and removal, good calibration with darks and kappa-sigma clipping should help.

Mark

[wimvb]

This is indeed walking noise and is caused by each frame a little different from the next. The best you can do to avoid this is:

1. make sure your rig is balanced

2. dither between exposures. Dithering (at least 12 pixels), effectively kills any pattern that hot/warm pixels can leave behind.

http://wimvberlo.blogspot.se/2016/07/dithering-in-hardware_7.html

 

In existing images, using an aggressive hot pixel filter together with darks during image calibration, can reduce the effect.

Cheers,

[matd]

Thanks for the replies. Yes I'm using kappa-sigma, and various combinations of calibration frames, but they don't seem to help that much.

It seems your'e right - it's also called 'correlated noise', there's a similar photo here:

http://www.cloudynights.com/topic/466251-correlated-noise-from-hell-last-night/

- it looks like I need to do several things;

  Use fewer, longer exposures (so there is less compounding noise) - I'll need remote software, as the in-camera max exposure is 30s.

  Get a light pollution filter so I can take longer subs without overexposing.

  Use dithering - for which I'll also need to use some remote-control software (if the Star Adventurer supports it). 

 

Nobody said it was going to be easy!

Cheers

-Mat

 

[matd]

* SOLVED *

- I'm an idiot, I was using the wrong set of darks (with a completely different exposure time). I noticed that without any calibration frames, the noise went away. With the correct darks, the stacked image looks fine and is ready for postprocessing:

Thanks for your help!

-Mat

[wimvb]

An alternative to camera control software is an intervalometer. They are quite cheap, available for most models dslr, and easy to use. In case you can't or don't want to have a computer out in the field.

this doesn't solve the dithering problem, of course

[SteveBz]

On 07/01/2017 at 10:25, matd said:

* SOLVED *

- I'm an idiot, I was using the wrong set of darks (with a completely different exposure time). I noticed that without any calibration frames, the noise went away. With the correct darks, the stacked image looks fine and is ready for postprocessing:

Thanks for your help!

-Mat

I'd love to see the end result.  I'm pretty much at the same place.  I find my stacking doesn't improve my image at all.  In fact the raw pictures seem to be marginally better than the stacked ones.  The only way I can remove them is by subtracting the noise and then it looks a lot better, but still not as good as those dark site images with stacking and heaps of post processing.

Here are before and after.  Like you I'm in a light polluted zone (near Gatwick Airport)

Regards

Steve.

[SteveBz]

On 07/01/2017 at 10:25, matd said:

* SOLVED *

- I'm an idiot, I was using the wrong set of darks (with a completely different exposure time). I noticed that without any calibration frames, the noise went away. With the correct darks, the stacked image looks fine and is ready for postprocessing:

Thanks for your help!

-Mat

Hi Mat,

I've reduced the noise in your photo, I just took the average of the first ten pixels and subtracted it from the remaining pixels.

 

[SteveBz]

I think the stacking has removed some of the detail in the right and upper right quadrants of your photo of Andromeda.  My single unstacked frame shows more detail, but it disappears on stacking.  However yours has more detail of the arms behind the main hub.

[matd]

Thanks for the replies. Here's my attempt at postprocessing it. Looking at it now, it seems a little false-looking.

I also need to sort out the vignetting (even though I used flat frames in DSS?).

I think I'll try it again and keep things more linear - I guess that would help make it look less overcooked?