TCampbell wrote in post #18317492
This is tough to process because it's difficult to determine what's real galaxy vs. sky background haze or light pollution.
I processed my own Andromeda in PixInsight. While PixInsight is a fantastic astrophotography image processing program, the one drawback is that 'masks' can only be generated by algorithm. You cannot using a mouse or graphics tablet & stylus to select a region. When I did mine, I had to produce a 'star mask' to isolate the stars and also a 'range selection' masks which uses relatively brightness intensities to try to identify the galaxy.
It's somewhat difficult to gradually increase the sensitivity of the tool to increase the range to try to get all of the galaxy to show up *without* starting to select sections of light pollution that shouldn't be selected (and the software doesn't give me the tools to manually reject parts of the mask that shouldn't be included -- the masks in PixInsight can only be generated via algorithm (you control only the parameters.))
The result (for me) was that I tend to be a tiny bit more conservative -- meaning I know my version of Andromeda doesn't quite extend all the way out to the edges of the disks. Basically... I know the challenges you would have faced trying to produce this image.
One astrophotographer friend did give me a tip to help deal with light pollution.
He mentioned that you can use the Digitized Sky Survey to download any section of sky you want. But the benefit of this is that these are images taken with sensitive cameras and in basically perfect conditions completely free of light pollution. So anything that gives off light in the image is actually in space and not a source from ground-based light pollution.
You can use the Digitized Sky Survey to align the sky survey to your own image (PixInsight can align any image even though it's not taken with the same scope/camera/resolution, etc.) and then used the Sky Survey to create a more accurate range mask to figure out where the *real* limits to the edge of the galaxy are found... and then use that to help you work your own image more accurately (you do not incorporate the DSS data into your image... you really only use it for reference purposes to help you figure out what light is "real" vs. what light is caused by local light pollution to help you get a more accurate mask for processing.)
I'm not quite sure what you're saying about light pollution with an image like this. The sky will be as dark as the light pollution will permit, and regardless of how bad it is, the outer haze of the galaxy will gradually fade into that light level. Light pollution will affect the entire area around a subject like this fairly uniformly - unless it's taken with a wide-angle lens that includes the sky over a town or other light source, and any difference between the surrounding sky and the hazy perimeter of the galaxy can only be caused by light from the galaxy. I don't think light pollution can contribute to what you see of the galaxy, only mask it and make it appear smaller than it would appear with no light pollution.
The pictures that were used to make up the above image of the Andromeda Galaxy were take from a Bortle 1-2 transition zone in the Ouachita Mountains in Oklahoma. There was very little light pollution there, and virtually no discoloration of the sky in any of the mages, even with a 10 minute exposure. It was DARK out there, and COLD, too. (16 - 18 Degrees), and the humidity was very low, so there was little to no haze in the atmosphere for distant lights to polllute.
This is a single frame from that session, right from the camera, without any enhancements at all, just reduced to post. For post-processing after stacking the 11 exposures, I used Photoshop and Irfanview. Total exposure time was 1 Hour and 10.4 Seconds.
This image was taken with a Nikon D5500, 55-200mm lens at 200mm, 299.9 Second exposure, f:5.6, ISO 1000. The White Balance was set to 'Sunny', and the Noise Reduction was 'Off'.
please log in to view hosted photos in full size.