Getting there Horse Head and M42

Peter.M · #1 17-10-2011, 02:20 PM

Hi guys, I have a mystery for someone. Thismorning I took some photos of the horse head nebula and the great orion nebula.

The orion nebula is 72 30 second subexposures at 1600 iso and 20 8 second mask frames.
The Horse Head nebula is 10 5 minute exposures at 400 iso.

To go with this I shot around 40 dark frames for the orion neb and 10 for the horse head. 26 flat frames were taken.

After processing there are weird radial noise lines through the images, I stretched the dark flat and bias frames and there is no sign of the noise. I then decided to stack the light frames with no calibration frames, that image did not have the lines but was badly vignetted and had far less detail.

I hope someone more experienced can tell me how to get rid of them as I should be imaging from a dark site this week.

Peter.M · #2 17-10-2011, 02:48 PM

For these images I have turned off the internal image noise reduction of the camera, I have a feeling that the lines are usually taken care of during this process. This saddens me a little because with the noise reduction on an exposure takes double the time. Oh well tomorrow I might try to do the same shots with noise reduction on and see what I get.

bmitchell82 (Brendan) · #3 17-10-2011, 09:05 PM

I used to get that noise from electrical interference. Be really vigilant on separation of your data and power.

Peter.M · #4 17-10-2011, 09:38 PM

I will do my best on that tonight, and try my luck at the orion neb all night I think. Im really keen to get something nice.

jenchris (Jennifer) · #5 17-10-2011, 11:09 PM

Darks are the same as the internal process of noise reduction.
I would have thought that it's not something you'd need to do.
I'm not sure RAW images have this done anyway - someone was telling me that only the JPGs get processed like that

bmitchell82 (Brendan) · #6 18-10-2011, 12:19 AM

For me as soon as I became anal about keeping my data and power apart the rainbow like noise ceased.

As for ICNR its alright while your learning how to take images but they are not good as the camera does some work on the image before you get it eg fiddling with the histograms which means you could be loosing data as astronomical work is normally working with very dull signal.

Peter.M · #7 18-10-2011, 03:15 AM

See, my thoughts on the issue were that if I was to take my own darks and then let dss subtract the master dark from the frames the result should be better, because the master dark is an average and therefore a better reprisentation of the noise in the image.

However I am currently taking more frames with noise reduction now on and I will see how it goes.

irwjager (Ivo) · #8 18-10-2011, 08:55 AM

Hi Peter,

I could be mistaken, but the artifacts in your image seem like the type of signature noise you'd get when you don't dither your subs.
Can you confirm there was no dithering between subs?

Cheers,

Ivo

Peter.M · #9 18-10-2011, 10:18 AM

Subs were dithered, and to my understanding dithering is only to eliminate hot or dead pixels. If you dont dither the hot or dead pixel will stay in the same place and will not be removed, If the frame moves slightly then data will be gathered for that pixel in subsequent frames.

irwjager (Ivo) · #10 18-10-2011, 10:46 AM

Quote:

Originally Posted by Peter.M

Subs were dithered, and to my understanding dithering is only to eliminate hot or dead pixels. If you dont dither the hot or dead pixel will stay in the same place and will not be removed, If the frame moves slightly then data will be gathered for that pixel in subsequent frames.

That's exactly right. Additionally it smoothes out any noise pattern that may be present caused by subtly hotter or colder pixels. Incorrect (or no) dithering can smear these patterns out in a nice line causing streaks.

Since you did everything right in the dithering department :thumsbup:, can you confirm your darks were shot at the same temperature?

EDIT: I know this is a beginners forum, but for more theory have a look at Mischa Schirmer's site - it's an awesome site with heaps of mathematically substantiated hints & tips. On dithering he says; if exposures are not dithered, then the noise contained in the dark frames is copied into the exposures and does not grow with sqrt(n), but linearly with n. This is because it is stacked on top of itself, whereas dithered exposures would lead to averaging out.

Peter.M · #11 18-10-2011, 11:30 AM

With noise reduction on and no dark frames I come up with this, these are 40 minutes of 3 minute subs. I think that i have stretched it too much but i am quite happy with it.

jenchris (Jennifer) · #12 18-10-2011, 11:38 AM

Where is the dithering option check box?

irwjager (Ivo) · #13 18-10-2011, 12:37 PM

Quote:

Originally Posted by jenchris

Where is the dithering option check box?

Your guiding solution (PHD, etc.) should give you the option to dither. If not (or you don't use auto guiding), just manually move the image a tiny bit between each sub, spiraling out from the position of your very first sub.

Quote:

With noise reduction on and no dark frames I come up with this, these are 40 minutes of 3 minute subs. I think that i have stretched it too much but i am quite happy with it.

Nothing wrong with that Peter! You did a great job keeping the core under control too.

Peter.M · #14 18-10-2011, 12:47 PM

Quote:

Nothing wrong with that Peter! You did a great job keeping the core under control too.

Thanks, I snuck in some cheeky 20 second subs for the core

Kirkus (Kirk) · #15 19-10-2011, 10:18 AM

Quote:

Originally Posted by Peter.M

With noise reduction on and no dark frames I come up with this, these are 40 minutes of 3 minute subs. I think that i have stretched it too much but i am quite happy with it.

The noise in your first 2 photos seem to be pretty uniform, almost like a star-trails shot. I would've thought electrical interference would be more random.

This second shot is beautiful though. I'd say whatever you did worked!