First (proper) go with LRGB Imaging

[Marky1973]

Earlier this week, I started out trying to get my first LRGB image since getting the ATIK 414EX and some LRGB filters. Wednesday night was clear and I got set up ready to have a go at a globular cluster - thought that might be an "easy" target to start with. And it was all going so well. PHD2 calibration and guiding was going smoothly - I could do no wrong!

 

Managed to get an hour of Luminance along with 15 minutes (2x2 binned) of RED GREEN and BLUE. Or so I though... after capturing the BLUE channel I realised I had forgotten to change the filter from GREEN- by which time it was too late and I had lost sight of my target (M13) so had no other option than to pack up and go to bed - lesson learned! :-)

I had a play with my luminance images (20x180 seconds) and struggled to get any colour with the RED AND GREEN channels, so just outputted a MONO image to see how it looked. This was processed in PIXINSIGHT - basic workflow (I'm still learning) with a CROP, DBE, HISTOGRAM STRETCH, HDR MULTISCALE TRANSFORMATION and then into PS for a quick tweek. I appreciate there is a lot more I could probably do, but I will get there eventually. I suspect I shoud dalso have included DARKS, but didn't this time around as I thought I could get away with it with the ATIK according to the manual. I did cool to -20, so hopefully there was little noise, but I can always take some DARKS and add them in later.

(The darker halos around the two larger stars will become apparent when we get to RGB'ing the image.)

I was fairly happy with that image for just an hour of data. The background felt like it could be a tad darker, but I was then losing some of the detail, so didn't want to push it too far. I guess more subs will help with this too. An hour is not that much in the overall scheme of things!

What I did want was to be able to add some BLUE to the image overall, so got out Friday night for a go at that. The forecast suggested a clear window of 2-3 hours around 11pm, so I should have been able to get 15 minutes easily. Mmmmm...if only the cloud hadn't rolled in.  My blue subs ended up a little poor because of the cloud, and the Halos around the larger stars are a result of this - their light was diffused around them because of the cloud - but I was determined to get some subs, so shot for an couple of hours and used the best ones.

But I ended up with 5 subs I could use, so back to PIXINSIGHT.

I stacked/calibrated/registered each channel separately, then used a similar workflow as above, until each image shared a similar size and resolution (as the RGB subs were 2x2 binned). I combined the RGB images into one and then used the LRGB Integration tool to apply the Luminance Channel to the RGB image.

Back to PS for a quick tweak and then saved a couple of tries.

 

A long way from perfect, but my main aim was to get some images and combine the LRGB channels - which I have managed to do. I suspect more and longer subs can only be a better thing! I am also sure that 15 minutes of binned subs are probably not enough to get a good image, so will be looking for more data next time out. One thing I noticed was the crosshairs on the two larger stars. Wasn't expecting this as I don't use a reflector. Is this just because they are much brighter stars, is this "bkloom"?

Overall, I am very happy with my first effort at LRGB. It is obviously a flawed image for the reasons discussed above but, with more subs and the slow development of my PIXINSIGHT skills, I am sure it is just going to get better and better.

The ATIK camera works really well with ARTEMIS and, despite a few quirks (that I am getting used to) it is very easy to work with - it does everything I need it to do at this stage in my imaging. I have used APT in the past, and the 414EX is compatible, but ARTEMIS also does pretty much what I need in a slightly more user friendly, less cluttered way. I'll keep hold of APT though as it is a great bit of software.

All in all, a good experience....now just keeping fingers crossed that Sunday night's clear forecast holds up!

[Uranium235]

Diffraction spikes on bright stars are completely normal - even with a refractor as they arise from the cell clips on your objetive lens. Take it as a sign you had it perfectly in focus... i'd be more worried if I didnt have them becuase it would mean something is off.

One thing to watch out for is your use of DBE, you must keep any sampling points well away from bright stars, background galaxies and your target - otherwise you will get "black holes"  in your images (which can be seen in the above luminance image). Also, as youve noticed - the blue channel suffered quite badly under poor skies (it always will) because blue light gets scattered by the atmospheric conditions.

Otherwise... not bad for your first stab

 

PS: Dont bother with darks... ever! (especially with your camera), as it will introduce noise into the image.

 

[Marky1973]

Cheers Rob, glad to hear I managed to get the focus right - seems so much easier with he ATIK/ARTEMIS and a Bhatinov mask - I still get a bit OCD about it, but it is much easier to see the effect of slight changes - much easier than with teh DSLR in LiveView on APT.

I know the picture isn't great, but my main aim was to see if I could at least combine the channels. Next step is to get much more data in each channel, perhaps without binning, over 4 nights and see how that turns out!

If the weather holds, I may get to start tonight, so time to think of a target.... :-)

I guess I won't bother with darks then.... are they any use with these cameras - perhaps on much longer exposures?

[Uranium235]

Darks? Nah, dont bother even with long exposures. I dont use them for my 383 either, which can deliver a snowstorm of hot pixels at times. I just stack em out with sigma clipping or use a kernel filter (in maxim).

[Pompey Monkey]

1 hour ago, Uranium235 said:

Darks? Nah, dont bother even with long exposures. I dont use them for my 383 either, which can deliver a snowstorm of hot pixels at times. I just stack em out with sigma clipping or use a kernel filter (in maxim).

Assuming, of course, that you're dithering between exposures!

I also use a KAF 8300 in my SBIG camera at -20 C. I used to only use a bad pixel map in conjunction with the DefectMap tool in PI. However, I found that using darks in the calibration too reduces my background noise a touch. 

[Uranium235]

No dithering here, never bothered and not likely to (too much of a faff for me)

[Marky1973]

I guess I can try some darks at some point - certainly less of a hassle now I can take them at any time with the camera cooling. Artemis doesn't dither from what I can gather (yet) so none here either. I was dithering in APT, but not using that with the ATIK (yet)... :-)

 

[Pompey Monkey]

Without dithering, how does sigma rejection of hot pixels happen?

[Uranium235]

Easy, there is a *tiny* bit of drift/flex in my guiding. Not enough to give mis-shapen stars or smearing in subs of up to 1800s, but enough to ensure hot pixels dont land in exactly the same place over the course of a session. I rarely go over 1200s, and im getting round stars - so I see no reason to fix it (and potentially break it)

 

[MattJenko]

I get exactly the same little spikes on my 414ex and ED80 combo with bright stars. I have done various experiments and the spikes are always vertical/horizontal and I put it down to the CCD window clips itself.

I also never bothered with darks for the 414. I did try but it seemed to add in noise! I use flats and a bias. Also, I do dither, but not sure how much difference it makes.