So much is lost in the down sizing.
Interestingly when the individual channels were stacked I used a feature in the software called "drizzle" which was originally written for the Hubble which makes better use of the data but the final stacked image before further processing is some
one and a half gigbytes , I end up with seven of these if I use all filters ( the above had no S11 or O11 by the way which may add something) but at that size the next bit of processing software really struggles often wont load and is ratty until I reduce size touchy touchy crash crash...which is a pain cause I crop at this point and the first sub gives a reference so a crash can see you start over when doing your fifth for example... I reduce each sub to at least half by a process called Binning which as I understand makes better use of the data whilst reducing the size.
Each of the subs are individually processed to make various improvements in the software used after the stacking software and among other things, reduces noise greatly...most effective really...signal to noise is what it's all about..like with the CBR map you will notice reference there to signal to noise concerns....
Anyways when your seven odd individual channels are processed you take them into photo shop RAW and have a bit of a play to tweak, exposure, contrast, dark highlights, light highlights, etc...then take them into the main section and give each channel it's colour, remember up to this point all channels are just mono ( black and white). When each has its colour that's when the fun starts because you are trying to balance to get colour correct but not lose detail....I usually duplicate each and can end up with twenty photos that I gradually merge I to one.
Finally I get them merged into one photo which I further process to adjust overall colour, brightness and contrast and highlighting of dark or light regions, and apply things like haze reduction. At this point I use the dodge tool to brighten certain spots , say the outter regions in the above photos to bring out the light nebulosity and use the burn tool to darken features say in the above to minimise the burn out of the gallaxy core...at this point I will create solid colour frames of a colour I wish to improve and layer the final photo over the solid colour and using the rubber tool let a slight amount of colour into the main image..I often give individual stars a little colour this way but it takes a long time as I may add a little red blue and yellow....so it's pretty well finished but I take that image into another program to adjust the brightness and contrast a little and even the levels miniscully, and down a down size there to be able to post it here say.
It is very time intensive as staking each sub frame can take hours x seven, each initial process is about one hour minimum x 7 and photo shop takes maybe 2 to 3 hours.
So as you can imagine this has not been in my days for six months ... usually I am processing day in day out.
Plus I have new software which
I have not really implemented that allows me to merge the data from two different rigs...so the above shots I can now merge for a better image.
Alex