Rho Ophiuchus Project (Work in Progress)

[Octane]

There's a first time for everything.

Inspired by Jase's mosaic presentation at AAIC 2011, as well as Rogelio's demonstration of PixInsight, I thought I would try and start to process some data that I've captured over the last month or so.

This composition is my second "proper" image ever taken with my SBIG CCD, and, the very first time I've ever indulged in a mosaic. I believe in diving into the deep end and learning how to swim.

Presented here is a rough-as-guts work in progress, consisting of six panels of luminance frames, totaling about 11 hours of data thus far. The goal is to capture a 4x2 mosaic, giving 8 panels and approximately 70 hours of exposure in total. I'm not sure if I will reach that goal this season, but, you can rest assured I'm trying to get out to dark skies every chance I get.

I am very pleased with my messing about with PixInsight over the last two days. It is a mammoth bit of software and its power is seen through pretty much every operation that you ask it to perform. I look forward to getting to grips with its minutiae that will hopefully allow me to create pleasing images.

Note: the panel in the bottom right sticks out like a sore thumb as it has less data than the other panels.

SBIG STL-11000M, Takahashi FSQ-106N 530mm at f/5
bin 1x1 -30 degrees

25% resolution available here (http://users.tpg.com.au/octane2/ro6.html).

Cheers for looking.

H

[strongmanmike]

Yeh but otherwise this is looking pretty damn good Humvee :thumbsup:

Mike

[dugnsuz]

H, it looks bloomin' awesome at this stage.
Amazing potential in this project.
Doug

[RobF]

Looking really nice so far Humayun. I would have expected PI to do better than that with the matching for the bottom right panel - did you have "frame adaption" turned on?

[DavidU]

It's going to be a blinder:thumbsup:

[richardo]

Really cool H... this is going to be specky when finished for sure.
Man you are a glutten for punishment...!!
But you're out and enjoying your nice setup, that's the main thing:)

You'll be waiting for the weather then you'll be waiting for the object again..:sadeyes:

Speaking of weather, haven't seen the sun in my region, let alone any stars for over a week now..... typical, have a few things to iron out and can't get out there... god I hate that!:rolleyes: aahh God u listening:question:

Good luck mate, hope you're having better luck than me!
Look forward to the finished article... then there's the colour:D

All the best
Rich

[alan meehan]

Looks good H its going to be good when its finished ,glad you are enjoying pixinsight its a wonderful program.
AL

[gregbradley]

Awesome start Humi. It looks fantastic.

You may be able to touch up those differences in brightness on the right side with the burn tool or the dodge tool in PS.

Greg.

[multiweb]

Pretty awesome field. Looking forward to the color. :thumbsup:

[RBA]

Looking pretty darn good so far!!

[Peter Ward]

I have no doubt the end result will be a masterpiece. Stick with it :)

[jjjnettie]

:) It's going to be your Opus H.

[suma126]

its going to look awesome when finished. great work

[marco]

Hi H, if "well begun is half done", then this is going to be an outstanding image :thumbsup::thumbsup::thumbsup:

I started a similar mosaic last year with the 300 EDIF (2x3 farames, including a wider area of course) and ended up in missing some color subframes here and there so now don't have all pieces of the mosaic (in particular about colors) to complete the big frame. Postponed to 201x when the 300 ED will be back on the mount :(

I also started another mammouth mosaic this year of the same area with the TEC :screwy:. It ended out that several subframes were trailed because of the problem reported in my recent ngc6188 post. Luckly the OTA is ok now and mounted so I hope to collect some more light and at least complete half of the mosaic originally planned :prey:

I have also purchased Pixinsight recently and I am experimenting with it these days. I would be very interested in sharing your experience using this program, in particular about mosaics..

Clear Skies
Marco

[Ross G]

Hi Humayun,

An excellent photo!

Super sharp, so much detail and what an amazing tonal range.


Thanks.

Ross.

[jase]

Maaate! Excellent work indeed H. You're well on your way to completing this project. Time to nail some chrominance data. I'm guessing you've now realised why in the presentation I have a few slides on data acquisition and planning as shooting RGB with a mono camera and filters, you need to step up the acquisition efficiency. Keep focus.

[peter_4059]

H,

I was going to say that looks awesome but everyone else already said that.

Looks like you are going to end up with a great image. Keep at it.

Peter

[RBA]

May I suggest a couple of things?

First, you may want to try the PixInsight's GradientsMergeMosaic tool developed by Georg Viehoever. More info, including how to download it (scroll down to Juan Conejero's second post) here: http://pixinsight.com/forum/index.php?topic=3211.0

I haven't tried that module yet, but it might be worth a try, and I will definitely be trying it for my next mosaic whenever that happens. It looks like Georg is definitely heading in the right direction with this module so as to make seamless mosaics a fairly easy task in PixInsight. If the "frame adaptation" option in StarAlignment was a great help, a tool allowing us to automatically correct for any other "deficiencies" is most certainly welcome!

The second approach is what I've done in the past sometimes, which, basically involves using the mask generated by StarAlignment and then run a DBE on the image with the mask activated, with a fair amount of samples around the conflicting areas. The DBE, in theory, would "soften" any brightness differences between the two frames, but because the mask is active, it would only be applied to one of the panes. This technique may require some practice with DBE, adjusting some of its parameters as well as placing/moving/deleting samples in a trial-and-error scenario, but in the end I've managed to get the job done many times - ok cross that "many" out.... some times :-)

In any case, having more/less data in a pane will always be problematic, because then, you're not just trying to get a linear fit between the two panes, but the different SNR may also become problematic (which may in fact confuse the linear fit done by the "frame adaptation" option). In that situation, you might just have to resort to continue processing with a non-linear stretch, and then make visual adjustments manually with the masks created by StarAlignment until things "look" right.

Hope that helps!
Rogelio

[RobF]

Good on you Humayun. Will be watching with interest how this one comes together. Thanks for putting it out there for all to see at such an early stage too. Another inspiring image for all us wannabes.

[Hagar]

Coming along nicely H. You better start praying to the weather Gods, you have a power of work to still capture your RGB.
Good luck with it.

[Octane]

Cheers, Mike. By the way, it was good to catch up with you again at AAIC. :)



Doug,

Thanks, mate. Here's hoping the potential is realised!



Rob,

Thanks! Nice to meet you up at AAIC, too. :)

I did have frame adaptation turned on, but, I'm not entirely sure what's happened with that particular panel.



David,

I hope so!



Rich,

I know, I'm a glutton for punishment. Not bad going for my second ever proper CCD image, right? :)

You're right, I enjoyed watching my guide star track properly and seeing nice round stars in each sub-frame as it downloaded was a real buzz. :D

I'm hoping I get some time away from work and the weather cooperates this week for some more data capture.



Cheers, all.

H

[Octane]

Thank you, Alan!

I'm not entirely sure why I had put PixInsight off for so long. I guess it was because I was a DSLR person for a while, and just got comfortable with doing things in IRIS. Seeing Rogelio's presentation at AAIC was the kick in the pants that I needed to start using the software. Very, very impressed thus far! It's not as scary as it all seems at first. I just started clicking on menu items and applying changes to duplicates of images until I got what I wanted. It's quite intuitive after a while; of course, I'm likely not using even 1% of the power and functionality of the software! All in good time. :D



Greg,

Thanks, mate.

I have a feeling that when I get more data on that panel and when the other two panels to the right of it are captured, that it should blend in quite well. Otherwise, as you suggest, fixing in Photoshop will be the order of the day.



Marc,

Thanks, dude. I'm looking forward to it, too. I really hope I get to finish this off this season. I'll be heartbroken if I don't get the opportunity!



Rogelio,

Thank you, so much. I must credit you for inspiring me with your presentation of the software; I'd likely not have been so keen to dive in otherwise. :)



Peter,

Thanks! It was lovely to meet you at the Malin's, too. :D



Cheers, all.

H

[Octane]

Jeanette,

Thanks, babez. Here's hoping. :)



Shane,

Thanks, mate!



Marco,

Thanks so much for the encouragement!

I have seen your Rho Ophiuchus on your web site every time I've gone looking at the masterpieces you post here. :)

Man, it's hard enough capturing this with an FSQ-106N, I can't imagine how much more difficult it would be with a longer focal length instrument -- yikes!

As mentioned re: PixInsight, I am a complete newbie; if I can get results like these after playing with the software for two days, then, you, no doubt, will master it quite easily and quickly!

I was really battling to stitch the images together. I had seen the video presentation on YouTube of the M31 mosaic being stitched together; in that video, everything just worked with default settings, if I'm not mistaken.

I would try and stitch two panels together and the software would just end up freezing on my MacBook Pro. I'd have to force quit the software. I then remembered that in Rogelio's presentation, he loaded his images up and then drew two preview boxes around the overlapping areas where the images would be stitched (make sure you click the use previews button in the StarAlignment mosaic registration routine). This, coupled with a RANSAC setting of about 4, worked flawlessly! I also had frame adaptation enabled to iron out the differences between the panels.

After this, I ran the DBE tool, then, HistogramTransferFunction for stretching. I also then ran a very mild MorphologicalTransformation to reduce the intensity of the smaller stars.

Mind you, this was a first pass -- once I have all my data captured and stitched appropriately, I am going to take a lot more care and a lot more time in doing these steps (and, whatever others may be required) to get the outcome I want.

Don't forget, there's always Photoshop to fix things in, too. :)

Hope this helps, at all? If you have any questions, re: settings/values/steps, I'd be happy to answer them.



Ross,

Thanks, sir. Glad you picked up on the tonal range. It's something I pay attention to in my landscape imagery, and, it just naturally comes across to astrophotography. :)



Jase,

Thank you, so much.

I'm hoping I get a chance this coming week to at least capture all the luminance data that I want (24 hours in total). After that, I'll set about capturing the colour data. Let's hope for clear weather.

As for capturing efficiency, I would be lost without PinPoint. Particularly as I haven't bothered getting my RoboFocus set up correctly, I'm re-focusing by hand (Bahtinov mask and software) every 30 minutes. PinPoint is my saviour, haha.

You are my inspiration. 'nuff said. :)



Cheers, all.

H

[Octane]

Peter,

Was lovely to meet you up at AAIC. :)

Thanks for the kind words. :)



Rogelio,

Awesome -- thanks so much for the GMM tool. Will check it out.

I have noted your masking/DBE suggestion and will attempt that if things go haywire.

As for not having equal amounts of data, I'm a real stickler for capturing exactly the amount of data that I initially intended to -- that is, 3 hours of luminance for each panel. The problematic panel here is just 1 hour of data, where, all the others are 2 hours.

I will keep experimenting. I'm learning quickly there's many ways to skin a cat in PixInsight. :D



Rob,

Thanks, again. :)

Mate, I'm on the learning path along with the rest of us.



Doug,

Thanks! I hear you, mate. I'm keeping a close eye on the weather this week. Must. Get. Data!



Thank you, everyone!

H

[Octane]

Grabbed the last two panels.

Reduction in Maxim DL and then mosaiced in PixInsight using the GradientsMergeMosaic module.

There has been no DynamicBackgroundExtraction or any other noise reduction other than a very, very quick histogram stretch (copied STF to HistogramTransformation) and a quick levels/curves iteration in Photoshop. I will need to play around a bit more to smooth it out, but, it is very, very promising thus far.

This is the first imaging I've done in six months. I am so glad that I still have the skills! Except for one newbie mistake -- I was wondering why my guidestar calibration kept failing on the STL-11000M in Maxim DL. Turns out I had forgot to plug the serial autoguider cable in. Oops! Once that was plugged in, it was smooth sailing.

9 hours in luminance and counting. I have 1 hour of red for panel 1 completed. I'll be heading out to Frank's every chance I get; just spent 4 nights out there but only one was useable until 4:30 AM. Friday night was perfect except for the gale force winds that were blowing. Didn't bother to image.

Click here (http://scratch.hqphotography.com.au/images/MergeMosaicCrop.jpg).

H

[troypiggo]

Promises to be an awesome project. I think we're just as excited as you to see the results :)

[naskies]

Very nice! :)

[SkyViking]

That looks spectacular H, and what a massive undertaking it is!
It also clearly gives the impression that the dust is lying in front of the background stars, very nice. Really looking forward to seeing this in colour.
BTW now I see why I couldn't get rid of the gradient in my recent M4 image, lol - it's covered in dust!

[astronobob]

struuth, amaizing what data can be achieved with diff techniques & equipment ! this is mindblowing out of my depth : ) amaizing to see the detail availiable & captured, priviledge to see it Humayun . . .

[dvj]

Excellent work so far. The final will be spectacular!

jg

[gregbradley]

Wow, that looks fantastic.

Greg.

[Martin Pugh]

Looks great H.

Martin

[RobF]

Your orbit looks good H. Keep bringing her in!
Inspiring as always. Was just lamenting on the weekend if you'd given up on the game.

I'm surprised gradient MM didn't even up the panels more though. Did you have frame adaption ticked when doing the Image Integration?

[seeker372011]

Frig...n' awesome

[TrevorW]

Awesome stuff H one of your best

[Octane]

Troy,

Let's hope so, mate. Weather permitting.



David,

Thanks!



Rolf,

Cheers, mate. I'm still daunted by the fact that I have to capture another 2 hours for each luminance panel, haha. In reality, by the time the mosaic is viewed at web resolution, no-one will be able to tell the difference between an SNR of 3 or 5 for the luminance data.

And, yes, it's quite dusty up there.



Bob,

Cheers, mate. Your version with camera lens was quite striking. Once I finish this monster, I will show you a comparison of what I captured 6 years ago (modified 350D and Nikkor 180mm f/2.8 ED lens) to what I capture today.



John,

Thank you, sir!



Cheers, guys.

H

[Octane]

Greg,

Thanks, mate!



Martin,

Cheers, sir. :)



Rob,

Slowly, but surely. :)

I have had thoughts about selling everything, but, I think those sorts of thoughts plague everyone who goes through dry spells, whether it be due to work commitments, personal life, or just the plain ol' horrible weather. I'm inspired once again and will continue chipping away at this one until I make the finest image of this region ever made. Maybe second to Rogelio's and Jase's efforts. I still go back and look at their work, as well as Loke Tan's image (http://sbig.impulse.net/userjpgs/iom/rho_ophiuchus.jpg) from many years ago (October, 2004) who was the original inspiration.

As for GMM not having done a stellar job, it is probably my fault as I'm a newb with PixInsight. I think I may need to run DBE on each pane prior to attempting to merge the mosaic. And, yes, I have had "frame adaptation" enabled throughout the effort.



Narayan,

Thanks! Good to hear from you. :)



Trevor,

Thanks, mate. I'm not sure where to go after this one. I do have a couple of mosaics planned (have had them planned since I bought the gear)... we'll see!



Thanks again, all.

H

[trent_julie]

H,
That's a wonderful shot so far, I am looking forward to seeing the final shot, I'm sure it will be magical!

Trent

[stevous67]

I think I'm never going to touch a mosaic project ever. Taking ONE panel properly is enough work, little alone perhaps 6-10. Then trying to seemlessly join them is all a little too much. But clearly H, your giving it a fair crack! Good luck.

Steve

[Octane]

Trent,

Thanks, mate. I'm also hanging for it, too. :)



Steve,

It definitely is a big task. But, then, you look at Rogelio's work and he's doing work with 54 panels! Granted, they're short exposures and only a few exposures for each panel, but, it still doesn't take away from the grandness of it all. My real love lies in long focal work (it's what I started off doing initially) and while widefields do appeal to me, and are challenging in their own right, I don't like them as much as long focal length images. So, to make things interesting (for me, and, eventually the viewer) mosaics are the best way to keep one busy and interested. :)



Cheers, boys.

H

[atalas]

H,looking good dude!hope you get the color soon....should be a ripper.

[marco]

Great preview H :thumbsup: I am sure this project will turn out majestic, it takes a bit of b@lls to start "proper" imaging (as per your statement) with such a big project ;)

I have a project running on the same area with 12 panes that started last year but have to say that I am unfortunately at a dead point at the moment, properly merge so many panes (in particular colors) is a very hard task, so you have all my appreciation in this one :thumbsup:

Clear skies
Marco

[gregbradley]

Hi H,

What is your workflow here? Do the luminance first and then align the rgb colour panels onto the luminance using Registar one at a time?

If you have the luminance stitched nicely then the colour may go on a lot easier as its more the gradients that seem to cause grief in mosaics.

PixInsight came out with a gradients for mosaics module some time ago. That may be very valuable for this kind of work.

Greg.

[Octane]

Louis,

Hey, cheers! Good to hear from you. Hope you've been making use of those dark NT skies.



Marco,

Thanks, sir.

I guess having such a wide field with the FSQ makes it hard to go after small objects (my real love is long focal length), so, got to mix it up and do something interesting. I look forward to purchasing a long focal length scope in the future.

I remember you mentioned ages ago that you had started on a mosaic but were having equipment issues, as well.



Hi Greg,

You're part-way correct -- no need for Registar. I do everything in PixInsight.

I took 8 reference luminance frames that were just 10 seconds long each, after having built the mosaic template in TheSky6. These are what I run MaxIm DL's PinPoint LE plugin on each time I go to start imaging. Sync the scope, re-goto centre of solved image. If it now looks bang on (or, close enough to being bang on), I start imaging.

For this iteration of the image, I took Rogelio's advice and tried the GradientsMergeMosaic module. Absolutely fantastic.

Mosaics are such a time consuming process that make you want to tear your hair out. I've only got 1 hour of red bin2x2 on panel 1. I don't know when I'm going to finish this project.

Still, something to look forward to.

Cheers.



Thanks, all.

H

[allan gould]

An impressive image to match your ambition, H.
Inspirational to say the least.