Reducing Blue Star Bloating from Doublet Refractor

[kayronjm]

Hello all,

I'm imaging with a Borg 77EDII refractor, QSI 660wsg-8 CCD camera and Astrodon E-Series Gen2 LRGB filters. As expected, I have some star bloating on the Luminance and Blue channels, though the bloating in the Blue channel is more difficult to deal with as it upsets star colours and the surrounding areas. I placed a Hutech IDAS LPS-P2 filter on the Borg 7870 focal reducer to help but it isn't perfect so I'd need to do some post-processing to help the problem.

I assume using StarMask to produce a star contours mask and then pinching the stars with MorphologicalTransformation (in Morphological Selection mode) is a good idea, but is there a better one?
I thought to do this to just the linear, stacked Blue image prior to colour combination with the Red and Green linear, stacked images.

Any better ideas are welcome. These are links to single exposures (10 minutes 1x1 of M101):

R = https://dl.dropboxusercontent.com/u/92999201/Borg%2077EDII/2016-06-04__M101_R_0600s_1x1_frame1.fit
G = https://dl.dropboxusercontent.com/u/92999201/Borg%2077EDII/2016-06-04__M101_G_0600s_1x1_frame1.fit
B = https://dl.dropboxusercontent.com/u/92999201/Borg%2077EDII/2016-06-04__M101_B_0600s_1x1_frame1.fit

Thanks!

[waynef]

I am also struggling with this.I tried starmask but when I stretch I just end up with dark halos around the stars, so I am missing something.Something that kind of worked for me was to desaturate in ps and then revert back and paint over the blue edges.But there must be a better true astro imaging process to get this correct.This only resolves the color issue though and not the bloating.Would be great if the experts can give their opinion

[msmythers]

Here is something I use if I don't get the focus just right on my AT72ED or my camera lenses. http://pixinsight.com/forum/index.php?topic=9250.msg59525#msg59525   It's two steps using RickS ColorMask tool. One is using MT for the brightness of the blue halo and then Color Saturation to reduce the blue itself.



Mike

[bmhjr]

Mike,  how do you determine the values for the hue in the color mask script?

Bill

[msmythers]

Bill

Read RickS's explanation just above my posting. http://pixinsight.com/forum/index.php?topic=9250.msg59395#msg59395 I would read the entire ColorMask posting also. http://pixinsight.com/forum/index.php?topic=7751.0

There have been some changes to the script since my posting but nothing that would keep anyone from using it the same basic way. The current version is v0.7 where I was using v0.4. You can see the differences in the attached image.


Mike

[RickS]

Mike,  how do you determine the values for the hue in the color mask script?

Bill: you can change the cursor readout mode to CIE L*c*h* and click around the image to get an idea of the range of hues in your halos.

Mike: glad to hear you're finding the script useful!

Cheers,
Rick.

[bmhjr]

Thanks Rick and Mike.  I knew there had to be a way to get the hue values, I just couldn't figure it out   

My modified DSLR produces red rings around some stars.  I tried using the script but it also selected the red nebula the way I was using it.  Hopefully with a narrower selection to the correct hue of the star I can get better results.  I will give it a try.

Thanks
Bill

[pfile]

you can also create a star mask, and then multiply the color mask by the star mask to get just the red star halos.

in pixelmath you'd say "$T*star_mask" and apply it to the color mask.

this would probably dim the color mask a little bit and give it the profile of the star mask. if you don't like that, you can say "iif(star_mask, $T, 0)" meaning if there is a non-zero pixel in star_mask then the corresponding location in the color mask should be preserved as-is, otherwise set to 0.

rob

[bmhjr]

Thanks Rob for the helpful suggestions.  I think one or a combination of these methods should get me there!

[bmhjr]

I used a combination of the suggestions
- Generate colormask script with narrow hue range from readout
- Generate Star Mask
- Pixel Math $T*star_mask applied to color mask
- Reduced saturation, it took multiple iterations
- MT to reduce stars a bit
- AT to sharpen stars back up a bit
- Final color saturation

Thanks for your help

[msmythers]

Looks nice.


Mike

[waynef]

These tips definitely are helping thank you everyone.
What I also discovered now that seems to work is to use curves with the starmask to tone down the out regions of the star.A very small contrast boost also helps.Then create a mask for just the outer halo and just the curves again to deal with dark halos by bringing the curve back up to match the background.Using the curves tool again with the same mask drop the saturation to match the background color.This removes some of the blue fringe nicely.

I just need to get better at making precise mask to mask the star and it halo.I always seem to end just over or under the halo of the star.Any ideas on how to get more accurate other than lots and lots of trail and error

[msmythers]

You might take a look at this posting about removing a circular reflection from Arcturus. It's a good technique for making a larger circular mask with a hole in the center. Gerald and others have a lot of good information throughout this thread.   
http://pixinsight.com/forum/index.php?topic=9141.0

Mike

[msmythers]

I almost forgot about this other posting about reducing 'multiple star halos'. http://pixinsight.com/forum/index.php?topic=9335    Raymond wrote a process for this case and later I made a video showing Raymond's work. So many ways of making masks and reducing halos for so many different situations.


Mike