First processed image (M42)

[CaptainShiznit]

Hey all,

So after finally getting a couple of hours of clear skies last night, I got to grips with BackyardEOS, set a plan and off I went. I live under terrible Bortle 8 skies so ended up going with 25 second exposures and I managed 97 before the clouds rolled in. I stacked in DSS using 50 Darks and Biases and 25 Flats and was left with the attached stacked image below. This was my first experience of real image editing (Photoshop) so I've tried to follow along some different tutorials and videos, the problem being the stacked images in the tutorials usually start out different than mine and so I find it tricky to follow exactly.

Regardless, I have a final image that I'm reasonably happy with for my first go but I'd welcome critical feedback. The reason I've included the original stack is in case anyone more seasoned in processing fancies a go. It would be interesting to see what else could be brought out of it.

Clear skies.

m42 stacked.TIF

[Anthony1979]

How did you processed it in ps... What tools did you use... Im having problems processing mine

[CaptainShiznit]

18 minutes ago, Anthony1979 said:

How did you processed it in ps... What tools did you use... Im having problems processing mine

Mostly just levels, curves and masks. 

This tutorial probably did the most for me.

[vlaiv]

Nice data, I believe there could be more in there - maybe if you post 32bit tif?

In any case - here is Gimp 2.10 quick processing:

I'm not sure that flat fielding worked as it should - there is still vignetting in that image. But on the bright side - Running man is visible as well and I think it can be pulled out with a bit more effort and 32bit tif version.

[Galen Gilmore]

Again nice data.

Here's my attempt with your image. I tried to give the image a more natural feel through color balance and contrast, although at the expense of smaller details.

[CaptainShiznit]

Nice work guys thank you.

10 hours ago, vlaiv said:

I'm not sure that flat fielding worked as it should - there is still vignetting in that image. 

I found the flats a bit tricky as it was my first attempt at calibration frames. I put a baby's muslin cloth folded once over the scope and secured with an elastic band then pointed at my laptop screen with a white browser page open. Camera was in Av mode. Do you have any suggestions for a better method?

10 hours ago, vlaiv said:

I believe there could be more in there - maybe if you post 32bit tif?

Again being my first attempt I just saved from DSS the default 16bit. What would be the better format between 32bit TIF/FITS and -interger/-rational?

[wimvb]

I just downloaded your image and opened it in PixInsight. What scope did you use?

8 minutes ago, CaptainShiznit said:

I put a baby's muslin cloth folded once over the scope and secured with an elastic band then pointed at my laptop screen with a white browser page open. Camera was in Av mode. Do you have any suggestions for a better method?

How large is your scope aperture compared to the screen? Make sure that the light source covers the whole aperture with room to spare. Also avoid too short exposure times. You may need to fold that cloth several times to get an exposure time that is long enough. With a mechanical shutter, you want an exposure time that keeps the whole shutter open long enough that the screen refreshes a few times. Alternatives to a computer screen is an LED panel.

Your image has running noise (or walking noise) in the background. While this can be held back during processing, it's better to avoid it all together by dithering during capture.

http://wimvberlo.blogspot.com/2016/07/dithering-in-hardware_7.html

http://wimvberlo.blogspot.com/2016/07/the-effect-of-dithering_21.html

[vlaiv]

23 minutes ago, CaptainShiznit said:

Again being my first attempt I just saved from DSS the default 16bit. What would be the better format between 32bit TIF/FITS and -interger/-rational?

Use 32bit float point. I prefer the use of Fits format as it is standard format for astronomy images, but many people prefer tif because it is easier to work with - most software for image processing will open that but not fits.

Gimp opens both. Do note that older versions of Photoshop can't work with 32bit data and you need 16bit data, and Gimp prior to 2.10 is limited to 8bit data only.

25 minutes ago, CaptainShiznit said:

found the flats a bit tricky as it was my first attempt at calibration frames. I put a baby's muslin cloth folded once over the scope and secured with an elastic band then pointed at my laptop screen with a white browser page open. Camera was in Av mode. Do you have any suggestions for a better method?

Wim gave better advice than I could because I don't work with DSLRs, but in case your flat calibration does not work - it could be due to darks rather than flats.

I would suggest that you use bias as master flat dark, use bias and darks and turn on dark optimization in DSS and then try calibration like that. While still not perfect, I think that would be the best way to calibrate your data.

[CaptainShiznit]

25 minutes ago, wimvb said:

What scope did you use?

SW 130 PDS. The laptop is a 19" screen and was maybe 2' away.

Thanks for the replies, fantastic advice as always 😊

[wimvb]

4 minutes ago, CaptainShiznit said:

SW 130 PDS.

Do you have a coma corrector with that scope? With a DSLR you really need one. If you already use it, you might want to check its distance from the chip, there is severe coma in the corners of your image.

[vlaiv]

Here is synthetic luminance from wiped channels in 1/4, 1/2, 1/4 ratio, stretched beyond breaking point - just to show that there is residual vignetting:

Cropped and a bit touched up:

It's actually quite decent

[wimvb]

Here's your image, processed in PixInsight.

• Gradient removal; as @vlaiv noted, there is vignetting in your image.
• extracted luminance

rgb:

• colour calibration
• chroma noise reduction
• arcsinh stretch
• histogram adjustments
• desaturated the background

l:

• noise reduction
• histogram stretch in multiple steps
• hdr compression
• curves transformation
• lrgb combination

I kept the background intentionally a bit dark in order to suppress the walking noise. With a dslr, dithering is your best friend.

Btw, since your gradient is uneven across the field, you may want to move the secondary mirror so that it is centred in the focus tube. If you like tinkering, that is.

 

This is the background of your image, as extracted by PixInsight, stretched

Edited January 28, 2020 by wimvb

[happy-kat]

I'm not sure the muslin fabric grain isn't being picked up there is a lot of streaky thin noise but I enjoyed processing it.

I used StarTools then irfanview to save the tif as png.

bin 50% | cropped the edges slightly | auto dev | wipe and used vignetting | manual dev | deconvolution | HDR on the core | colour | Psycho-Visual Grain Equalization De-Noise

I did a second process that was streched further and from that took the running man and blended it to the first image.

[CaptainShiznit]

1 hour ago, wimvb said:

Do you have a coma corrector with that scope? With a DSLR you really need one. If you already use it, you might want to check its distance from the chip, there is severe coma in the corners of your image.

No I don't yet but it's on the list 😅 I was prioritising an LP filter first.

24 minutes ago, happy-kat said:

I'm not sure the muslin fabric grain isn't being picked up there is a lot of streaky thin noise

I'll try a t shirt next time as I generally see recommended. I just went with the cloth as it was to hand 😊

 

These are some beautiful images you guys are producing. I really hope I can get close to this level of wizardry before long!

[ollypenrice]

This was with a quick version of my basic routine, so DBE and SCNR green in PI then over to Photoshop for the rest. I rotated the image till the background streaks were vertical then ran Noel's Actions Reduce Vertical Banding to reduce that aspect of the noise. I used a basic log stretch in Levels and a minor further adjustment in Curves then used the colour select tool to reduce colour saturation and colour noise in the background. I boosted the colour using soft light as blend mode on a top layer and applied it as a colour layer to the original. I also went into Lab colour mode and increased contrast on a and b channels. (Both of these techniques are explained in Rob Gendler's 'Lessons from the masters.')

It's very decent data. Longer subs would go deeper, that's all. Oh, there was a curious artifact just above and right of the Trapezium. Maybe a flats defect or an internal reflection but I repaired it by eye (some call that painting! Shame on me!!)

Olly