Getting "Hubble Palette" style images from your OSC dual/tri narrowband data - PixInsight Tutorial - UPDATED

[Luke Newbould]

Hi guys, I made a quick video tutorial to try and show an easy to follow method to take your OSC narrowband data taken with filters such as the l-extreme and other similar ones and turn it into a hubble-esque palette.

I know that there's lots of ways people like to perform this sort of task, but this is how I've self-taught and I thought if I can do it then anyone can!

 

This tutorial uses PixInsight, the EZ-processing suite, available free here https://darkarchon.internet-box.ch:8443/ and also finally requires that you have Starnet installed and working in PixInsight, though I believe this is standard for new versions.

 

I hope this is useful for some of you!

 

EDIT - I made available a newer version of this tutorial which is in my opinion better - I'll leave the old one here for anyone interested still though!

 

the older one is below -

 

Edited June 11, 2021 by Luke Newbould
Added a video revision

[Budgie1]

Thanks Luke,

Got some hand tips from the video.

Nicely explained and easy to follow.

[Lee_P]

Great video, thanks for making it

 

[Luke Newbould]

Thanks for the lovely comments guys! I'm really happy it was of some use 😊

[scotty38]

Nice one good work. I don’t have PIxinsight and didn’t bother pursuing after the trial as I didn’t really get my head round it but this is making me reconsider....

Edited March 14, 2021 by scotty38

[Luke Newbould]

@scotty38 - I had a rocky and frustrating start with PixInsight for the first three weeks or so of the trial, and I seriously thought I'd not be purchasing it, especially given the lofty price tag! - Things started to click into place though soon afterwards, and now I do genuinely enjoy processing in it

 

[Chefgage]

Great tutorial. Seriously thinking of upgrading my current processing software to pixInsight now that I am imaging more useful data.

[Craney]

Excellent video Luke.

I have always been reluctance to embrace PixInsite, feeling I needed a higher degree in image processing just to crop the raw image.

This has eased the divide and I may start to invest time in it.  I will check out your YT channel and see if you have other great tips and guidance.

Cheers,

Sean.

 

[Luke Newbould]

Hey @Chefgage and @Craney! - that's awesome you are liking the look of it, I couldn't really go back from processing in PixInsight now i'm over the initial struggles of getting started in it - it's transformed my imaging!

Since the video I posted I've done another couple ones using similar techniques, just really showing that it works on a variety of nebulae and not just something tremendously strong like the Rosette - they might be of use to you if you are interested!

 

Thanks again for checking it out!

[Chefgage]

4 hours ago, Luke Newbould said:

Hey @Chefgage and @Craney! - that's awesome you are liking the look of it, I couldn't really go back from processing in PixInsight now i'm over the initial struggles of getting started in it - it's transformed my imaging!

Since the video I posted I've done another couple ones using similar techniques, just really showing that it works on a variety of nebulae and not just something tremendously strong like the Rosette - they might be of use to you if you are interested!

 

Thanks again for checking it out!

I will go and have a watch. Thanks for taking the time to produce these tutorial type videos, it really does help.

[Luke Newbould]

@Chefgage Thank you! I hope it's useful

 

I actually just today will be releasing a better made tutorial, better audio quality and steady pacing (I feel) - It's scheduled to be available from 1pm onward if you have any interest in it mate!

 

Good luck! 👍

Edited May 27, 2021 by Luke Newbould
message body edit

[Anne S]

I watched your video this morning and then reprocessed a close up of the centre of the Heart nebula. Imaged with an elderly Atik320e in April, so not much data even though it was over 3 nights. 900s subs with a L'enhance filter. I was never particularly happy with my original processing. The Hubble style version shows much more detail.

I'll be looking at some of your other videos. Thank you for putting in the time to do them.

 

[Luke Newbould]

Great job! @Anne S  I'm glad to see it's worked well with an older camera too

I hope you enjoy watching some other videos! I appreciate that a lot, thanks! 👍

[Lee_P]

Thanks again for making the tutorial. It's pitched at a perfect level for me (and we have the same camera and filter!) I used your tips to help edit this Elephant Trunk photo. Amazing what we can do nowadays with OSC cameras even in light-polluted skies (I'm in Bortle 8). I'm definitely interested in any more videos you make! 

 

[Luke Newbould]

@Lee_P Thank you Lee! - I'm so happy to have helped you in some way mate - you really have made a beautiful image there, and genuinely I think anyone looking at it would have an incredibly hard/nigh on impossible time determining if it had been done with a mono camera or not, which really highlights what a great job you did with your data, and the power of these new OSC cameras/filters + the processing options we have available these days!

Bravo!! 👍

[tomato]

Hi Luke, 

Thanks for your tutorial, I had a measly one hour of of data on the Elephant's Trunk region, captured with a RASA8 and QHY268 OSC using the IDAS NBX filter and got a 50 shades of salmon pink image.  Following your mk 1 tutorial I have managed to get something a bit more Hubblesque. It needs more work as the dark regions have lost detail but thanks for taking the time to make the video share it.

 

 

[Luke Newbould]

@tomato - You are most welcome! - That's a great shot mate! - Thank you for sharing that comparison, you've got some lovely cold looking blues to come out in that pool of OIII, very nice job already I'd say!

[tomato]

Hi Luke,

Inspired by the first result I put some salmon pink data from the Heart Nebula through your version 2 tutorial. I can safely say this is the most processing I have ever done in PI, albeit religiously following your script.

Please excuse the dodgy blue band at the top, I was too idle aligning the camera on successive nights, and the star shapes are nothing to write home about either, but the end result does please me more than my original.👍

[Luke Newbould]

@tomato Hey mate! another lovely job, well done indeed!

I know what you mean about the standard salmon pink look most of this type of data takes on, which can still look effective on the right target of course! - That said, on most targets I prefer a colourful approach, like what you have done here!

Keep it up mate, you have really gotten the hang of it!

[Sterrenland]

Just watched the video...although I've only just acquired a dedicated astro camera (ASI294MC-Pro) I'd love to try this in the future. Found the video very well-explained and inspiring!

Out of interest is the Optolong L-Extreme one of the better filters to try with a OSC camera to replicate the Hubble effect?

[Luke Newbould]

@SterrenlandThank you! I really appreciate the feedback I've tried it out with l-enhance data and l-extreme data, the l-extreme does work better, but both are viable!  I'd imagine many of the duo/tri narrowband filters on the market will work similarly well

 

I hope that helps answer your question a bit 👍

[tomato]

I use the IDAS NBZ dual band filter with my OSC camera, and can confirm the  data works fine with Luke’s workflow.👍

[Lee_P]

I reprocessed some old data using Luke's technique. Happy to get some blue into my L-Extreme data!

 

[Luke Newbould]

@Lee_PPhwooaaarr, phenomenal result!! really well done mate!! 🤯 I'm loving seeing the processes, thanks for sharing!

[Lee_P]

Hi @Luke Newbould, I'm reprocessing another image using your method. I've made a slight change, and am wondering whether this it's something you didn't do on purpose -- if not, it might be a neat optimisation. Basically, I'm at the stage where you've split the channels into R G B. You've deleted B, and are using Curves and HistogramTransformation to make G look like R. What I'm currently trying instead to achieve this is to open up LinearFit, use R as the reference, and then apply to G. I *think* that achieves what you're aiming for with the Curves / HistorgramTransformation steps, but is simpler. Or maybe I'm missing something! Whaddya reckon?

Thanks,

-Lee