A Few Firsts; LRGB, Bi-Colour and Ha - Cygnus Loop and Iris

[Hytham]

Good evening!

I'm relatively new to the world of imaging (started one year ago), new to this board and I wanted to share some of my CCD firsts (about 2 months now): Cygnus Loop in Ha, Cygnus Loop in bi-colour (Ha + OIII) and a widefield image of the Cepheus region highlighting the Iris nebula with a couple of LBN objects.

All images were taken with a Tak FSQ106 EDXIII, FLI ML16803 and Astrodon filters at native F5. Image acquisition details are found in the Astrobin descriptions, but please ask if you want more info.

I'm very open to criticisms as they only help me improve. So please don't be shy if you don't care for an image and telling me why (I have very thick skin).

Thanks for looking.

Cygnus Loop Ha Small:

Cygnus Loop Bi-Colour Small:

Widefield Iris (NGC 7023) LRGB Small:

-Hytham

[Guest]

Cygnus B in colour is just beautiful, would be fantastic mounted on a wall!

[Deezel]

Fantastic images Hytham!

Really like the way you pulled the dust out of the Iris.

Is there a tutorial on how you did that with Pixel Math anywhere?

[Martin-Devon]

Super images, for somebody that is only 1 year into imaging, these are outstanding. As above, I think the Iris shot is rather special since you've pulled a lot of dusty nebulosity out even with relatively limited exposure times. Excellent.

Martin

[JohnC64]

fabulous images

[ollypenrice]

This is great imaging. Well done!

Are you not tempted to add some RGB to the Veil in order to get some star colour? The depth is fabulous and the processing superb.

Fabulous dust in the Iris, too. If I were going to be picky (and at this level of high quality I'm inclined to be picky!) I'd just say that I think the Iris core is a little blown and maybe colour saturated?

If this is after only a year I have no idea where you'll be in five years...

Olly

[iksose7]

You wont hear any complaints from me Hytham, those are truly stunning images! The wide field Iris is especially impressive to me, that dust is making me salivate

[Hytham]

Thank you, everyone. I really appreciate all the compliments and criticisms! Only way to learn, right?

Deezel, I taught myself by reading the pixel math documentation and using some of the more famous individual's cheat sheets - being a techie kinda helped me pick up on the programming behind it all. I can write something up for you here when the work day is over and you'll be surprised at how relatively easy it is. Unbelievably powerful utility. Of course with anything, I'm sure there is more than one way to skin this cat - and easier.

Olly, thanks! I've followed your images for a long while now and have always been impressed. So thank you for the compliments!

I do want to add RGB data to the Cygnus Loop, but my excitement kinda pulled me away into other projects (such as the Iris). With that said, what kind of exposures would you recommend to collect the RGB data for the stars? I was thinking 3x3 bin 2x2 for each wavelength to overcome the shot noise in such a small capture sequence.

I agree that the Iris core is blown out and I *think* I know where it happened. When I was re-applying the enhanced dusty image back to the original, I may have forgotten to use the specific mask for the Iris to protect it during the multiple iterations OR it was when I used an exponential transform in pixel math prior to applying back to original. I'm betting on the former.

I also agree the blue is a bit over saturated and I admittedly did that on purpose. I wanted it to stand out amongst all the dust, but I would say you're right. I'll dial it back.

Thanks again everyone.

[andyo]

Very nice indeed

[ollypenrice]

Thank you, everyone. I really appreciate all the compliments and criticisms! Only way to learn, right?

Deezel, I taught myself by reading the pixel math documentation and using some of the more famous individual's cheat sheets - being a techie kinda helped me pick up on the programming behind it all. I can write something up for you here when the work day is over and you'll be surprised at how relatively easy it is. Unbelievably powerful utility. Of course with anything, I'm sure there is more than one way to skin this cat - and easier.

Olly, thanks! I've followed your images for a long while now and have always been impressed. So thank you for the compliments!

I do want to add RGB data to the Cygnus Loop, but my excitement kinda pulled me away into other projects (such as the Iris). With that said, what kind of exposures would you recommend to collect the RGB data for the stars? I was thinking 3x3 bin 2x2 for each wavelength to overcome the shot noise in such a small capture sequence.

I agree that the Iris core is blown out and I *think* I know where it happened. When I was re-applying the enhanced dusty image back to the original, I may have forgotten to use the specific mask for the Iris to protect it during the multiple iterations OR it was when I used an exponential transform in pixel math prior to applying back to original. I'm betting on the former.

I also agree the blue is a bit over saturated and I admittedly did that on purpose. I wanted it to stand out amongst all the dust, but I would say you're right. I'll dial it back.

Thanks again everyone.

RGB for the Veil complex; you have two possible objectives, I think. One is just to get star colour to add to what you have. This would not take long but, whatever you do, don't bin 2x2. This will give you nasty big blocky stars which will not sit well with your tiny tight NB stars. Besides, stars are bright and so don't have any trouble overwhelming the noise. At the focal length of the FSQ I wouldn't bin ever, under any circumstances, whatever the image or filter. For star colour you could do half an hour per filter in 5 minute subs. You'd be laughing with that.

The other option is to try to get real RGB signal throughout the image. HaO111O111 does do a good job but it does not have the subtle colour range/depth of RGB enhanced by NB in the appropriate channels. If you want to go for the best possible RGB contribution then several hours per channel would be nice!! I'd go for 10 minute subs or even 15. How long is your patience?

I'm still faffing around with my Veil and getting it a bit better, but I didn't shoot any luminance, just Ha, O111 and RGB. I don't think lum would be useful but, of course, I might be quite wrong about that. Experimentation is the only way...

Olly

[Hytham]

RGB for the Veil complex; you have two possible objectives, I think. One is just to get star colour to add to what you have. This would not take long but, whatever you do, don't bin 2x2. This will give you nasty big blocky stars which will not sit well with your tiny tight NB stars. Besides, stars are bright and so don't have any trouble overwhelming the noise. At the focal length of the FSQ I wouldn't bin ever, under any circumstances, whatever the image or filter. For star colour you could do half an hour per filter in 5 minute subs. You'd be laughing with that.

The other option is to try to get real RGB signal throughout the image. HaO111O111 does do a good job but it does not have the subtle colour range/depth of RGB enhanced by NB in the appropriate channels. If you want to go for the best possible RGB contribution then several hours per channel would be nice!! I'd go for 10 minute subs or even 15. How long is your patience?

I'm still faffing around with my Veil and getting it a bit better, but I didn't shoot any luminance, just Ha, O111 and RGB. I don't think lum would be useful but, of course, I might be quite wrong about that. Experimentation is the only way...

Olly

Great points and thanks.

Regarding your comment about block like stars ... that is exactly what I ran into when binning 2x2 within my Iris Nebula image. I had thought it was one of three things: 1) The 2x2 binning, 2) The Star Alignment process in PI, or 3) A combination of points (1) and (2). Your comment is exactly what I had hoped to be confirmed.

Thankfully in my line of work patience is a must and I have plenty of it. I will be spending about 4 days in Northern Minnesota camping and will have my equipment with me under some very dark skies (Bortle blue). I'm currently working on 1 project (Heart and Soul, with Double Cluster HaLRGB Mosaic) and going to get a head start on another (Perseus to Hyades Mosaic also in HaLRGB). If time allows, I'll definitely grab at the very least 1 hour in each broadband wavelength.

[Hytham]

Hey Deezel,

As promised here is part (1) of the quick and dirty tutorial on using pixel math to extract the dust. If you have questions, please feel free to yell.

I'm only covering the equation used, and will not cover star reduction, and masking techniques. Please keep in mind that what you see below is highly exaggerated because none of the masks have been applied or created. I quickly recreated the process to at least get you started.

Pixel Math equation:

1 - ((1-$T)^2)

^^Take the bright and dark portions of the image and inverses them proportionally

I'll break it down into segments so that it is easier to understand:

"1": is a representation of of the white point of your image. If you were to open up a non-linear image in PI, run pixel math and enter the value "1" into "RGB/K" field, uncheck the "Rescale result" option, check "Create new image" and select "F5", you will have yourself a newly created white image.

"$T": represents the target image

"-" and "*": are pretty self explanatory; subtraction and multiplication

"(<expr>)": anything falling in between the "()" means the expression is evaluated with the highest precedence

"(<expr>)^2": Takes the result of the expression to the power of 2.

Knowing that any expression falling in between the "()" takes precedence, we know that the first step in the equation is the evaluation of "(1-$T)^2" which squares the subtraction of the target image from the white image. We then take the resulting image from the equation and subtract it further from "1" aka the white image. The result of this equation is the intermediate image we'll use to further enhance all of the dust in our main target image.

Looking at the result of each part of the equation:

Star reduced image (you must first reduce the stars in the image, and then create masks to protect the very bright areas prior to applying the equation above):

(1-$T) produces:

(1-$T)^2 produces (as you can see there is a lot of dust):

1 - ((1-$T)^2) produces the final intermediate image we will use to combine with our original:

-Hytham

[Deezel]

Thank's a million, Hytham!! I really appreciate you taking the time to do this.

I've been meaning to use Pixel Math for other processes than just narrowband mapping, now I have no excuses.

[WaveSoarer]

All of those images are really stunning.