Horsehead from Bromley

[Carole]

Rather disappointed in this but I think I have been too spoilt by going to DSCs. 

Luminance was pathetic and RGB was not much better, and only salvaged by adding Ha to the Red channel.  I haven't even used the luminance it was that bad.  I didn't take the LRGB until after the Moon had set and spent most of the evening, doing Ha.  Even Mark's Arcsinh process did not work on this.

Ha 21 x 900secs + 10 x 600secs - Almost 7 hours
RGB 32 x 150 binned total 1hour 20mins
Grand total 8 1/4 hours
SW130PDS & Atik460EX, HEQ5

This is the Ha only version.


This is the rubbish HaRGB version.  Think I am going to have to do the LRGB at a dark location if I can get to one in the winter.



Not a patch on Duncan's image.

Carole

[MarkS]

Hi Carole,

I don't think the RGB version is rubbish at all - it looks pretty good to me.  Obviously Duncan has the advantage because of his dark sky but your version is excellent for the light polluted suburbs.  It will be very difficult to bring out the dust with a good dark sky.

Mark

[Carole]

Sorry Mark, I think I misled you, as what I labelled the RGB version, should have said, "This is the HaRGB version".

These are the individual channels, only given a couple of iterations of levels and then curves and a couple of iterations of despeckle.  Red I guess is not too bad, but green and blue are pretty dire.

[MarkS]

You asked what are the swathes of fuzzy stuff.  It's impossible to say.  They are gradients that need to be removed.  They could be caused by optical effects including reflections within the scope or it could be passing cloud during imaging. You are trying to extract very low signal levels out of a strong light pollution gradient.  The tiniest variations in that gradient are then magnified by a huge amount.  At a minimum you need very carefully created flats but even then the problem will still be there.

Light pollution gradients are also the main limitation on my own imaging.  When I try to extract extremely faint dust from my images I find gradients all over the place.  It's exactly the same issue - the data must be stretched by an enormous amount to show the faint dust but this also stretches the tiny variations in gradient by an equal amount.

Mark

[Carole]

You are absolutely right Mark, I have had another process of the RGB data and did exactly that, using selective gradient exterminator and I think I have made a better job of it this time.

I have posted it in another thread as well.



Carole