Why so blue?

[Ivor]

I've found some time at last to start processing some subs I took back in January, the image is 40 X 15 sec Darks, Lights and 30 Flats & Bias. I've only stacked the subs in DSS but the image appears very blue, any suggestions why?

Subs taken on unmodified EOS400D

[RobertM]

Hi Ivor,

If you haven't colour balanced it then explains the result so do that first.  Also since it's an unmodified camera then you'll find it a bit light on Ha of which this nebula does have a fair amount.

Robert

[Ivor]

Cheers

How do I decide the best colour balance for my camera? I know I can push the red high in levels in photoshop but that seems a little too inaccurate and relies on previous knowledge?

[The Thing]

Hi Ivor,
Nice effort, better than my first attempts at M42. Longer subs will help with the colours.

Also make sure that DSS is set to not use the camera white balance or automatic white balance. The option is in the last section on the left on the RAW tab. This preserves what was captured (should be .CR2 RAW files not .JPGs). You can do a fair job of balancing the colours in DSS by making sure the peaks of the histogram line up. You can also do a lot of stretching using the sliders.

[MarkS]

Ivor,

Nice image. 
The data looks fine and you haven't burnt out the Trapezium.
Once the colour is balanced it will look really good.  The red (H-alpha) will definitely require quite a lot of boosting because of the (unmodified) camera's insensitivity.  As for colour "accuracy", what do you want to look accurate, the star colours or the nebulosity?  Every CCD has a different spectral sensitivity and certainly different to that of the human eye, so any talk of "accuracy" doesn't really make a lot of sense.  Do what I do, and simply adjust until it looks "right"!

Mark

[Ivor]

I feel I'm near an eureka moment (however that could be the bottle and half of red), as I think I'm nearly there but I'm missing a step at the start. I been saving my subs as FITS but couldn't see an option to disable the auto balance (is this an error in my choice of file type?) Equally I have my camera set to AWB so is this where I going wrong from the start?


The DSS processed result has the channels aligned so should I be stretching the red more in FITS liberator to compensate?

[The Thing]

Ivor,

AWB is fine, FITS is fine. Check colour settings in "Options, Settings, RAW/FITS DDP Settings" bottom left. This needs to be in effect before you process subs.

Move the Luminance sliders so that the colour peaks are around the middle of the S curve, then go to the RGB Levels tab and get the colour peaks to line up ( you will have to go back and forwards a few times) the up the Saturation slider. You will be amazed at what you can pull out of the image just using DSS.

BTW I think you will find the image is more grey than blue, upping the saturation should reveal the true colour cast. Tweaking the sliders will bring out more colour.

Duncan

Note:
1) when you process the first time a set of Master Offset, Dark and Bias TIFs are created. Uncheck all these types of subs and add in the Masters in thier place, then DSS doesn't process the files every time, much quicker.
2) when trying settings find a nice sub, then select a rectangle on it (Region of Interest). DSS will process this part of every image which is far quicker than doing the whole thing. The deselect the area when you have the settings right.

[Ivor]

Thanks Duncan

My setting for DSS FITS is as follows:



Anything wrong?

I can see by making making the adjustments you suggest the images has potential however I get black marks on the images which implies I've overcooked it.



In an effort to be more mathmatical and to stretch and scale the image I like to use FITS Liberator, I normally use ArcSinH(x) however in the case I've tried most of them and I end up with the black marks as well..




Any sugguestions on how to prevent the black marks?

[The Thing]

HI Ivor, you should use AHD debayering, and I think you need to adjust the kappa value of the sigma kappa stacking to get rid of the marks.

[Ivor]

Thanks I shall change it to AHD. I currently have the Kappa value set to 5 what would you recommend?

[The Thing]

5 sounds a bit high, I usually use around 2 or less in Deep Sky Stacker, 0.8 has given me good results. I usually only use one iteration as well. If you use a high Kappa value not much will be rejected.

"Kappa-Sigma Clipping
This method is used to reject deviant pixels iteratively.
Two parameters are used: the number of iterations and the standard deviation multiplier used (Kappa).
For each iteration, the mean and standard deviation (Sigma) of the pixels in the stack are computed.
Each pixel which value is farthest from the mean than more than Kappa * Sigma is rejected.
The mean of the remaining pixels in the stack is computed for each pixel."
http://deepskystacker.free.fr/english/technical.htm