M17 (WiP)

[lazjen (Chris)]

This (portion of) M17 is a Work In Progress that I hope to update over time as I learn more processing with PixInsight. The aim is to make a decent image from the data using PI only.

I've got this data, taken with a STX8E:

Temp: -15
6 x 600sec Ha
6 x 600sec SII
6 x 600sec OIII

Darks: 12
No Flats or Bias

Eventually I'd like to mosaic the region, but for now, I'll just use this data.

This version is a beginning and I used PI for most of it with a final cleanup in PS.



Astrobin: http://astrob.in/118616/0/

In PI, I did:
* Made a master dark.
* BatchPreProcessing script to calibrate and register the images.
* ImageIntegration to create Ha, OIII and SII images from the calibrated, registered images.
* LRGBCombination on SII, Ha, OIII as R,G,B respectively.
* SCNR to reduce Green
* BackgroundNeutralization
* ColorCalibration
* MaskedStretch
* HistogramTransformation
* Deconvolution
* DynamicCrop

That's for this particular image. I did quite a lot of messing around with various Processes after the "ImageIntegration" above. My main focus with this first attempt was to try to get the colour right, which I did not succeed in doing yet. This means, for example, the stars need work (especially the 2 brightest ones). With my other attempts, I found I was often wrecking the colour balance significantly and getting magenta stars and/or washed out backgrounds, etc.

Now that I've had a good play with these processes (and more), I think I'll revisit the tutorials and videos to see what I should be doing - they might make more sense this time.

However, any hints, tips, etc would still be appreciated!

[SimmoW (SIMON)]

Nice, but still looks very monochrome? Not that I'm an expert....

Thanks for posting, as a PI user I wasn't aware of the Masked Stretch tool, looks good to retain star colours!

[lazjen (Chris)]

The monochrome look is probably because of my attempts to fix the green colour, plus a lack of saturation on the output.. I guess.

[aarong (Aaron)]

Great start. I've just taken the plunge with PixInsight so seeing your workflow is really interesting. You can increase the saturation via the CurvesTransformation tool. Click on the S on the bottom-left under the curve and move the curve to adjust the saturation.

There are also useful scripts for combining narrow-band data that are worth playing with.

I have found Harry's astro shed a good place to start for LRGB images, but there is lots to learn. I'm still working out the best workflow for my images.

Cheers,
Aaron

[lazjen (Chris)]

Thanks - yeah it one of my many stuffing around attempts I discovered the saturation button there.

And as you mention, there's scripts for combining data - just discovered those this afternoon. In my few attempts I didn't find the SHO-AIP one to be that useful.

I have to go back to Harry's site (btw, your link has an error in it) and watch some more videos.

This afternoon, I started working through this tutorial:http://pixinsight.com.ar/en/info/pro...or-helmet.html I haven't completed it yet, but so far I've got an image with (mostly) better nebula but the stars are super-wrecked magenta and red.



The flow is mostly as per the tutorial up to just after the two MaskedScripts. After that, I messed around a bit (sat, etc) because I felt as if I'd gone off the rails (and dinner approaches).

[RickS (Rick)]

Good start on narrowband processing, Chris! There's lots of ways to skin this cat and it's a matter of finding the ones that work for you.

One suggestion: you need to do deconvolution while the data is still linear, at least if you care about the mathematical correctness of the operation.

Cheers,
Rick.

[lazjen (Chris)]

Thanks Rick. Potential dumb question time: how do I know when I'm going non-linear? What operations do I have to be cautious about?

[RickS (Rick)]

Chris,

The data is linear so long as it is linearly proportional to the original signal. That relationship is destroyed when you perform a non-linear stretch on the data, typically with HistogramTransfer or MaskedStretch.

Deconvolution and ColorCalibration should always be done with linear data. Some processes can be done on linear or non-linear data (e.g. noise reduction with MLT/MMT/ATWT/TGVDenoise, albeit carefully if the data is linear) and a lot of other processes work best on non-linear data, e.g. HDRMT and LHE.

Cheers,
Rick.

[lazjen (Chris)]

I've watched some of Harry's videos and have made another version.

Key aspects in this version are a LinearFit on the channels to reduce some of the colour imbalance; BackgroundNeutralisation; an attempt at Deconvolution (while linear) using a star mask (this bit I intend revisiting as I haven't got it as good as it could be); HistogramTransformations; HDRMultiscaleTranform (with a luminance mask); ACDNR (this didn't seem to do much); CurvesTransformation for saturation.

Star colours are just wild, but I see that others haven't bothered trying to correct star colours either e.g. http://www.astrobin.com/13254/0/



Astrobin: http://astrob.in/118616/B/

Another area I'm warring with myself is the background. In some of my versions I've got a lot more colour that I think I might be removing too much. Compare the above image to the first version posted (original post). I see more detail in this one, but far less colour. If I try to keep the colour, I get a lot more appearing than in that first version, e.g. this:



Astrobin: http://astrob.in/118616/C/

The two bright stars on the right continue to be difficult to deal with - especially the double cross-spikes they've both got.

[RickS (Rick)]

Looking better, Chris, and you've got to play with a bunch of different tools.

Stars colours are difficult with narrowband. Typical strategies are (a) desaturate them all to white (b) replace with RGB stars (c) just ignore them

Tweaking colour and contrast is tricky and relies a lot on personal taste.

Cheers,
Rick.

[lazjen (Chris)]

I like option (c) but I'll try for option (a) at some point. I think I just can't live with them being so red.

I'm going to pause on this project for bit and work on some other data I think, then revisit it for another attempt.

[RickS (Rick)]

Troy Piggins has some PixelMath that can help with magenta stars: http://www.iceinspace.com.au/forum/s...d.php?t=124463

For M17 I found it worked OK but missed a few stars that were more on the red side of magenta. The ColorSaturation process is another option for this.

Cheers,
Rick.

[lazjen (Chris)]

After a few weeks break and working on some other material, I thought I'd have another go at this data.

I went back to scratch and instead of using the BatchPreprocessing script to calibrate, register, etc., I did it myself one step at a time. This gave me what I consider to be better calibrated frames for the steps that followed.

I did four versions today as I worked through various paths and attempts. Core to all attempts were the use of LinearFit before colour calibration and the PixelMath from Troy Piggins that Rick mentions in the previous post. I also used HDRMultiscaleTransform for all as well.

On the last two versions, I used deconvolution with a star mask protection to sharpen the image more. And as I did it, I was aware of Rick's warning, but chose to ignore it this time.

The final image is here: http://astrob.in/118616/H/ The other versions can be seen there as well (version E, F, G).

[RickS (Rick)]

Getting there, Chris

[lazjen (Chris)]

Indeed. As I was writing this up for G+, I realised another set of alternate workflow steps I can use in conjunction with today's efforts. So, there might be more versions yet.

[lazjen (Chris)]

Another day, another attempt: http://astrob.in/118616/I/

As I mentioned in my previous post I thought of other things I could try. While it didn't quite work out as well as I would like, I think this is a better version, especially for the nebula details.

[LightningNZ (Cam)]

That's looking really good Chris.

[lazjen (Chris)]

Thanks Cam - maybe one day I can add more to it and expand it further.

[Rex]

Looking good Chris.

[lazjen (Chris)]

Thanks Rex - it gives me something to do while I wait for a night to get out again. Very annoying at the moment, it's clear skies but blowing dreadfully.