View Full Version here: : M20 with QHY8
Hagar
15-05-2008, 10:49 AM
Another go at M20. 10 X 5 min + 10 X 10 min no darks etc.
iceman
15-05-2008, 10:52 AM
Wow Doug, lots of exposure which is great.
However on the processing side it looks very clipped on the black point, over-saturated on the nebula and too much smoothing/noise reduction as well.
Can you post a stacked image with no processing?
Are you not taking flats yet?
Wow 2.5 hours of exposure !!!
Well done, but I have to agree with Mike, the image looks clipped and too much smoothing/noise reduction applied.
With so much exposure you wouldn't need to apply so much n/r or smoothing.
Good effort Doug. 2.5 hours should deliver a good result. A few comments;
The image looks heavily sharpened/deconvoluted due to the dark ring around the bright stars. You've either pushed this too hard by mistake or it was intentional. If it was intentional i.e. you wanted to sharpen other aspects of the image, you should layer/mask the stars to keep the star aesthetics.
As the others have mentioned the data is black clipped, so you've lost data in the fainter regions. The histogram shows two peaks which isn't normal. On closer inspection, it is the green and blue channels which are clipped. The red looks fine. This is the cause of the two peaked histogram. (See attachment below).
Due to the RGB misalignment your colour balance is not correct, but weighted in the Red. On reflection, you've got close to the right tones for this target as there is still quite a lot of overlap between all three channels - presented in the blue box on the attachment.I think some more time reprocessing this target would see a considerable improvement. You've got the data, so you should be able to stretch it rather hard without introducing considerable noise. I would suggest focusing on the 10x10min exposures in the first instance. Leave the 10x5min data for star colour management if needed or if there is an area of the nebulosity that is blown or washed out (white clipped). Even with the 10x10min data you can still manage the vast dynamic range of a target by stretching it at different intensities, then re-layering as required. Don't combine the 10x10 and 10x5 into a single image as this does not give you the best control in handling the range. Work methodically on a single set of data first.
Look forward to seeing more.
Hagar
15-05-2008, 01:36 PM
Thanks Guys I will give this another go. The joy of procesing when nearly asleep. Night shift and imaging don't go hand in hand.
Thanks Jase for the comments. I will give it another go with just the 10 minute data and see where I end up. This is all new to me. Your comments are valued.
Lester
15-05-2008, 02:20 PM
That image has got great potential Doug, looking forward to your repro.
It is still very stunning with the vibrant colours, your camera is working very well.
Hagar
15-05-2008, 02:46 PM
Another try at processing. This is only the 600 second data. I think it is better. Please let me know.
iceman
15-05-2008, 02:47 PM
Loads better, Doug! Very nice indeed.
There's an overall magenta/blue colour cast now but again, just processing and can be reduced/eliminated next repro :)
Hagar
15-05-2008, 03:04 PM
Another small change to the colour cast.
Lester
15-05-2008, 03:17 PM
Look at the nebulosity there, very nice Doug.
Just needs a tad more blue and less purple IMO.
Tell you what Doug, for someone doing image processing for a short time, your doing very well.
I guess its from the help everyone is giving.
Theo.
monoxide
15-05-2008, 04:16 PM
looking good Doug, getting some good data there.
vBulletin® v3.8.7, Copyright ©2000-2025, vBulletin Solutions, Inc.