Pics from 26-02-09

bluescope · #1 01-03-2009, 09:00 PM

Hi All

Here are some images I did a couple of nights ago plus a repro of Omega Centauri from earlier in the month. When processing the three images from 26-02-09 I discovered radial noise patterns on the data. Has anyone any experience with this sort of thing and what may have caused it. It's very annoying ... made processing a lot harder ... in fact probably screwed the images all together as it was imposible to remove effectively and the images are softer than they could be. The other thing about the noise was that with M83, a shorter exposure, there was less of this effect.

Any clues IIS GURUS ?

ic434 and ngc5128 are both 10x360 sec
m83 was 5x360 sec

renormalised (Carl) · #2 01-03-2009, 11:55 PM

Despite the noise, they're good piccies. Well done

bluescope · #3 02-03-2009, 02:33 AM

Quote:

Originally Posted by renormalised

Despite the noise, they're good piccies. Well done

Thanks Carl !

Tom Davis (Tom) · #4 02-03-2009, 02:36 AM

They all look good to me. Yes, the background noise can distract but I still think more data stacked would clear up the noise. Make sure you dither between frames though. This will help you put the noise pattern in a offset from frame to frame. That way it will mean or sigma out better.

http://www.hiddenloft.com/notes/dithering.htm
THis example (from 2003) uses dithering in CCDSoft, but Maxim and many other control programs can dither now.
Tom

bluescope · #5 02-03-2009, 03:00 AM

Quote:

Originally Posted by Tom Davis

They all look good to me. Yes, the background noise can distract but I still think more data stacked would clear up the noise. Make sure you dither between frames though. This will help you put the noise pattern in a offset from frame to frame. That way it will mean or sigma out better.

http://www.hiddenloft.com/notes/dithering.htm
THis example (from 2003) uses dithering in CCDSoft, but Maxim and many other control programs can dither now.
Tom

Thanks for the link Tom ... I'll have to read through it ! Glad you like the pics anyway.

renormalised (Carl) · #6 02-03-2009, 08:56 AM

Quote:

Originally Posted by Tom Davis

...Make sure you dither between frames though...

Tom

Tom, tell Steve to do that and he'll get nothing done!!!!

peeb61 (Paul) · #7 02-03-2009, 04:53 PM

A fine bag of images there Steve!
Very nice indeed.

Paul

rogerg (Roger) · #8 02-03-2009, 05:14 PM

Nice array of pics, despite the noise.

I see the pattern in the noise that you're talking about. I usually get patterns in noise when the dark frame(s) I'm using are incorrect in some way. Perhaps one dark frame for the whole night, and the camera varied temperature, so then some noise is left and other pixels subtracted, combined with some drifting in images (perhaps field rotation having an effect) ... I can imagine those kinds of factors leading to it. This would especially make sense if you were guiding on a star which is at the centre of the circle made by the ark of noise.

Short of changes to dark frames used, I would process out the noise some more by darkenning the background (and yes, perhaps losing some brightness in the objects themselves in the process) and perhaps running medians on specific colour channels to see if one in particular is a culprit.

Half this might be irrelevant depending on your camera, I can't remember what camera it is that you're using sorry.

Anyhow, nice images, just not perfect, but hey, can't be too picky

Roger.

bluescope · #9 02-03-2009, 08:01 PM

Quote:

Originally Posted by peeb61

A fine bag of images there Steve!
Very nice indeed.

Paul

Thanks Paul, glad you like them.

Quote:

Originally Posted by rogerg

Nice array of pics, despite the noise.

I see the pattern in the noise that you're talking about. I usually get patterns in noise when the dark frame(s) I'm using are incorrect in some way. Perhaps one dark frame for the whole night, and the camera varied temperature, so then some noise is left and other pixels subtracted, combined with some drifting in images (perhaps field rotation having an effect) ... I can imagine those kinds of factors leading to it. This would especially make sense if you were guiding on a star which is at the centre of the circle made by the ark of noise.

Short of changes to dark frames used, I would process out the noise some more by darkenning the background (and yes, perhaps losing some brightness in the objects themselves in the process) and perhaps running medians on specific colour channels to see if one in particular is a culprit.

Half this might be irrelevant depending on your camera, I can't remember what camera it is that you're using sorry.

Anyhow, nice images, just not perfect, but hey, can't be too picky

Roger.

Hi Roger

I always do darks to match particular image sets so I don't think it's due to that. I also did specific noise reduction filtering on individual channels with not a lot of improvement as the noise patterns were quite large. Anyway I'm still quite happy with the images ... as you say they're just not perfect ... and perfect is what we all strive for I think.

I use an SBIG ST2000XCM.

Thanks for your thoughts on the noise problem mate !

strongmanmike (Michael) · #10 02-03-2009, 09:12 PM

Quote:

Originally Posted by bluescope

Anyway I'm still quite happy with the images ... as you say they're just not perfect ...

maybe not but they are very very good

I too think that is noise from under exposure at high temp, take heaps more data.

Mike

bluescope · #11 02-03-2009, 09:33 PM

Quote:

Originally Posted by strongmanmike

maybe not but they are very very good

I too think that is noise from under exposure at high temp, take heaps more data.

Mike

It was a pretty warm night ... I was down to about -12 to start and about -15 by the end of the session. Definitely more data would help to smooth it all out.

Looking forward to cooler weather and some of my favourite targets.

strongmanmike (Michael) · #12 02-03-2009, 09:46 PM

Ok I didn't know you were cooled...? Hmm?...still looks like underexposure noise like you can get with Halpha filters on faint nebulosity if the subs aren't long enough but they aren't short exposures really and they were at pretty low temp

Is this what you always get?

Mike

bluescope · #13 02-03-2009, 10:16 PM

Quote:

Originally Posted by strongmanmike

Ok I didn't know you were cooled...? Hmm?...still looks like underexposure noise like you can get with Halpha filters on faint nebulosity if the subs aren't long enough but they aren't short exposures really and they were at pretty low temp

Is this what you always get?

Mike

No ... I've only noticed it on these images

bluescope · #14 03-03-2009, 12:36 AM

Just did a bit of a tweak on the three new images. When I looked at them again they seemed to be a bit heavy on the green side of things so I've cut it back and also made them a little smaller 900x675 instead of 1200x900.

multiweb (Marc) · #15 03-03-2009, 05:55 PM

Cool pics. The two galaxies look very good.

bluescope · #16 03-03-2009, 06:47 PM

Quote:

Originally Posted by multiweb

Cool pics. The two galaxies look very good.

Thanks Marc !