IC410. Adding HA to the RED channel or LUM of RGB images.

A place to share images that you have taken with SharpCap.
Forum rules
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.

Please share the equipment used and if possible camera settings to help others.
Post Reply
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

IC410. Adding HA to the RED channel or LUM of RGB images.

#1

Post by timh »

Here H alpha data were used to enhance the luminance, the red channel or both of an RGB OSC image of the IC410 tadpoles.

Predictably, the luminance enhanced image shows much better detail and depth in the nebula while the red channel-enhanced image shows less detail but much better colour ; clearly there wasn't enough chrominance for the enhanced luminance level. Lesson learned here I think (and based on median image intensities) was that it would take at least another 6h of RGB imaging (and/or alternatively of OIII imaging) to better balance the 2h of HA capture and produce an image that would be easier to process with both good colour and deep nebula detail. Not really surprising probably given that, compared to the mono camera, less than a third of the light is going into each channel of the OSC camera.

Image A is the RGB image (background extracted (DBE, colour corrected (Photometric col cal)
Image B is RGB with about 0.5 of the (HA - red channel) 'starless' difference image added to the red channel (PI pixmath)
Image C is RGB with the RGB L channel Linear fitted to the HA, a new L channel created as the MAX(HA, RGB L) and then replaced into the RGB image
Image D is C but with a slight further addition of HA into the red channel.

Personally I think that Image B was the best compromise and that is shown in full. Quite pleased with it but the road to perfection is certainly endless! At least by analyzing results you can at least see the next steps.

All above operations were carried out on Linear images and then PixInsight further used to reduce noise (MLT) and to stretch (Curves).

Bortle 6 skies (with no moonlight for the RGB images)

Skywatcher 200 PDS Newtonian, Baader flattener, F 5.0, CEM70 mount with PDS guiding via AS 120 MM camera and 9 x 50 finder. Pegasus Cube focuser

ASI 294 MM PRO camera, -10C, Optolong 7 nm HA filter
ASI 294 MC PRO camera, - 10C, using an Astronomik UHC filter

Image scale was about 1 pixel/ arcsecond with FWHMs of stacked images about 2.8-3.2

Sharpcap livestack captures filtered on FWHM and Brightness, Master darks (33-40 frames) and flats (100 frames) captured in Sharpcap.

124 x 70s (145 min) of RGB at gain 124 and 38 x 3 min (108 min) of HA at gain 124 captured by SC and preprocessed in PI



TimH
Attachments
tadpoles (2).JPG
tadpoles (2).JPG (110.37 KiB) Viewed 1266 times
IC410_NEWT_Baader_070321_RGB_124x70s_gain124_pix_integration_ABE_SCNR_PCCcol_PIXMATH_PLUS 38 x 3min HA gain 124 _280221_MLT_starmask_Trans_curves_curves_px_small.jpg
IC410_NEWT_Baader_070321_RGB_124x70s_gain124_pix_integration_ABE_SCNR_PCCcol_PIXMATH_PLUS 38 x 3min HA gain 124 _280221_MLT_starmask_Trans_curves_curves_px_small.jpg (446.16 KiB) Viewed 1266 times
User avatar
Menno555
Posts: 1053
Joined: Mon Apr 20, 2020 2:19 pm
Location: The Netherlands
Contact:

Re: IC410. Adding HA to the RED channel or LUM of RGB images.

#2

Post by Menno555 »

Very cool this Tim!
The road to perfection indeed has no end ;)
For me Picture C somehow has the most "pleasing" view, don't know why though.

Menno
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: IC410. Adding HA to the RED channel or LUM of RGB images.

#3

Post by oopfan »

Hi Tim,

Great work!

Hopefully you don't mind but I downloaded your image and annotated it. I identified two stars in Panel C, labeled "I" and "II". Using C2A, I looked up their Color Index:

Star "I": Color Index 10.32 - 9.169 = 1.151
Star "II": Color Index: 11.132 - 10.693 = 0.439

The Color Index of a Sun-like star is approximately 0.6. Red stars have a Color Index greater than 0.6, and blue stars are less than 0.6.
Therefore, Star "I" is red and Star "II" is blue.

To my old eyes, Panel A and Panel C are the best at reproducing star color. Although I love the detail in Panel C, my vote goes to Panel A.

After viewing tons of narrowband images of this nebula, I remind myself of this truth:
"Hydrogen Tadpoles swimming in a Sea of Oxygen".

Panel A shows Hydrogen Tadpoles (red) swimming in an Oxygen Sea (cyan).
In terms of signal strength, OIII is weaker than Ha, so my recommendation is to capture several hours of OIII.
The tricky part is how to mix the OIII into the nebula without skewing the stars??

Brian
timh tadpoles.JPG
timh tadpoles.JPG (176.06 KiB) Viewed 1245 times
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Re: IC410. Adding HA to the RED channel or LUM of RGB images.

#4

Post by timh »

Thanks Brian,

You did something there that I don't know how to do yet which is how to submit an image, plate solve it on line and then map the relevant annotation. I must learn that trick. It is a useful one.

It makes sense that the colour should be more accurate in A and C as you have confirmed. I am learning that there is really no 'free lunch' when adding extra narrow-band signal to any one of the channels. It is always going to distort the original RGB image in some way. Some of the inherent challenges seem to be ...

1) The red channel subtraction from the HA signal never works cleanly - for example the HA signal is in many cases overall the stronger but the HA stars tend to be relatively smaller and less bright - and alignments never quite perfect. Mathematically the difference signal is anyway always going to be noisier and comprise more artifacts than either original (the same is true for O3 signal plus blue and green channels but noisier).

2) For big emission nebulae the added HA difference signal ends up the major part of the total red channel signal. This then seems to cause the red channel to 're-normalize' versus the fixed RGB luminance level and compress downward the red signal of the underlying RGB image- as indicated I think by the observation that the background sky moves towards a blue/ green colour whenever lots of HA is added to the red channel.

3) The alternative of adding the HA signal to the luminance channel does, as you observed, retain the original colour but it becomes too thinly spread versus the luminance with a consequent pallid appearance and increase in chrominance noise.

So once you have added HA to the red channel I think that the final colour inevitably all ends up rather arbitrary anyway. For image C I just played with the colour settings until it looked 'OK' - as if it were a true RGB image. You are right though - one way forward would be to supply those tadpoles with some oxygen to balance things up and there is indeed OIII signal around the tadpoles. I have about 2h of OIII signal so far but am going to need at least another 4 h and meanwhile the skies aren't cooperating while IC410 is ever disappearing into the west.

TimH
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: IC410. Adding HA to the RED channel or LUM of RGB images.

#5

Post by oopfan »

Hi Tim,

It's not a plate solve. I used the features of C2A, planetarium software (free). First step is to search for "IC 410", as you can see in the attached screenshot that I annotated. The red box on the left are photometric magnitudes. These are available in the UCAC4 catalog. When you first download C2A, it comes with the SAO catalog. You need to manually download the UCAC4 catalog. At least that is the way it was 5 years ago. The author may have streamlined it since then. The user interface is configured so that the SAO catalog stars are shown when you zoom out past a particular threshold. And when you zoom in, it will switch from SAO to UCAC4. The first two magnitudes in the info list are "B" and "V", respectively. There are a wide variety of photometric filters, but the "Johnson-Cousins" system is widely used. A "B" filter is deep blue when you hold it up to white light. A "V" filter is light green. The "Color Index" of a star is the difference between the B mag and the V mag, specifically B-V. A Sun-like star has a Color Index of about 0.6. So called "Carbon Stars" are deep red. They have Color Index of about +3. Some hot blue stars have a negative Color Index. Generally speaking, I regard stars over 0.6 to be red, and under 0.6 to be blue.
timh_color-index_C2A.jpg
timh_color-index_C2A.jpg (310.15 KiB) Viewed 1225 times
With regards to preserving accurate star color while using narrowband filters, maybe there is a way to create two layers. One layer is reserved for stars that you isolate using "Sky Net" or "Star Net" or whatever it is called. The other layer is for the nebula. The star layer is fed data from just RGB filters plus true, wideband Luminance. The nebula layer is fed data from your narrowband filters. I've never tried this, but in theory it seems reasonable.

Brian
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: IC410. Adding HA to the RED channel or LUM of RGB images.

#6

Post by turfpit »

Tim

Online plate solving http://nova.astrometry.net/ see the Upload tab.

Dave
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Re: IC410. Adding HA to the RED channel or LUM of RGB images.

#7

Post by timh »

Many thanks Brian and Dave!

Both useful things to know to identify what you are looking at. I will have a go at identifying some of the faint fuzzies I often see in the background as well as the more obviously coloured stars.

Tim
Post Reply