I've been trying to image the Tarantula for a long time, with the DSLR it was always green with more stars than nebulosity. Now, with the QHY8 we see a bit more Ha and a slightly redder spider.
There are so many stars in the field, the nebulosity gets lost a bit. Perhaps I need to capture some Ha data to emphasise the clouds a bit more. I've seen some excellent narrowband images on this forum and I'd like to give it a go myself sometime.
This image is a stack of 20x5min subs, QHY8 and 127mm APO, guided with a QHY5+ED80+PHD on an EQ6 EQMOD Ascom pulseguide. Levels, curves, unsharp mask PS CS3.
Very nice and sharp shot Rob. How did you combine the subs? SUM or MEAN? I found mean to keep star colors better and to be softer on the stars when stacking. Try different way of combining the data and it might soften the whole look.
Hi Rob. This is a very nice image. The QHY8 does a great job on targets such as this and you have captured it very well.
I had a little play with your image. I hope you don't mind. All I did was stretch a little more of the red channel in curves, dropped the brightness a touch and moved the histogram a little more to the left and balanced the colours using levels. The white or blue white stars are very bright and make adjusting them on this image imposible. You may want to look closely at the way you apply curves and keep the brightness down a bit.
Hope this helps.
Hi Rob. This is a very nice image. The QHY8 does a great job on targets such as this and you have captured it very well.
I had a little play with your image. I hope you don't mind. All I did was stretch a little more of the red channel in curves, dropped the brightness a touch and moved the histogram a little more to the left and balanced the colours using levels. The white or blue white stars are very bright and make adjusting them on this image imposible. You may want to look closely at the way you apply curves and keep the brightness down a bit.
Hope this helps.
Doug, thanks so much for the advice, of course I didn't mind you having a play with the image. You made such an improvement it's hard to believe it's the same image Your advice is very useful, I've had a quick go at the original image and the improvement is amazing. You've led me down a new track. More to follow.
Quote:
Originally Posted by seeker372011
The Kappa Sigma stacking with default settings I find usually works well for me..average is not the greatest
Hi Narayan, I just tried Kappa Sigma with the same subs and I think the result is better. Processing using Doug's advice should produce a better image. I did find that average worked better for my Horsehead though. I need to learn more about how these stacking methods actually work
Hi Rob. This is a very nice image. The QHY8 does a great job on targets such as this and you have captured it very well.
I had a little play with your image. I hope you don't mind. All I did was stretch a little more of the red channel in curves, dropped the brightness a touch and moved the histogram a little more to the left and balanced the colours using levels. The white or blue white stars are very bright and make adjusting them on this image imposible. You may want to look closely at the way you apply curves and keep the brightness down a bit.
Hope this helps.
I shot a wide field of the same object at http://www.multiweb.com.au/astro/NGC2070_pano_s.jpg
The colors I got in the photo were the exact same tint as Rob's original photo.
I'm just wondering how we got the color wrong. Does the color balance change with every different DSO you shoot?
I shot a wide field of the same object at http://www.multiweb.com.au/astro/NGC2070_pano_s.jpg
The colors I got in the photo were the exact same tint as Rob's original photo.
I'm just wondering how we got the color wrong. Does the color balance change with every different DSO you shoot?
I wonder... The Tarantula is mostly Ha I believe, which should appear in the red channel, but with I see a lot of data in all 3 channels. Does this mean the green and blue pixels are sensitive to Ha?
I wonder... The Tarantula is mostly Ha I believe, which should appear in the red channel, but with I see a lot of data in all 3 channels. Does this mean the green and blue pixels are sensitive to Ha?
Hi Rob, my understanding is that Halpha is deep red as in M42 or the red part of M20. Hbeta is the bluehish part of M20. In both shots we got on NGC2070 if I check my color balance I have all the RG&B histograms aligned and equivalent in width/size. I suspect that if you check yours they'll be the same. Having said that every single "original" shot I got of the Tarentula with the QHY8 has this green tint. Like yours.
I agree if you offset and stretch the red channel then you get what Doug ended up with. But is this the correct color? I mean if we applied the same stretching to our M42 shots we'd end up with a M42 that would be too red right?
Hmm, I looked it up. NGC2070 emits light in many frequencies, it is a violent and complex region. http://www.eso.org/public/outreach/p...hot-14-02.html
From this page:
"A large number of different and colourful objects are seen in this amazing image. The very complex nebulosity is prominent in most of the field; it predominantly emits red light from hydrogen atoms (the H-alpha spectral line at wavelength 656.2 nm) and green-blue light from hydrogen atoms (H-beta line at 486.2 nm) and oxygen ions (two [O III] lines at 495.7 and 500.7 nm)."
So it would seem, a neutral colour may actually be correct. Although I must admit I prefer the red version
The reprocessed image is really lovely. In comparing the light I get on my mono images, the green and blue filters predominate near the centers of nebulae where you would expect excitement of Hydrogen at the beta wavelengths plus some blue reflection depending on the nebula. Further out, red predominates giving way to the ultra red Ha which picks up just about all the visible features because its emitting everywhere. Yes Ha is deep deep red but rendering it that way 'looks' bizarre and also loses all the beautiful luminance it can provide so it seems that most of our ha rendering is a compromise between colour and luminance. Rob, I think you are right that in terms of 'real colour' a neutral image is probably closer to being correct as our eye isn't selective for 656.2 nm.
Thank you all for the positive comments.
The support and advice we all get from the IIS community is beyond value. I for one feel that my skills are improved every time I post an image here.
I have only put up an alternative to the original image. A few things worth remembering about astro imaging and in particular the QHY8 are The QHY8 has a sensitivity towards green, having twice as many green pixels in the Bayer grid as the other two colours. Ha is predominantly an orange red colour. The incidence of a reflection nebula and the presence of quite heavy dust clouds around Tarantula will in my opinion at least, tend to darken both the reflection nebula and the emmission nebula quite some.
The other thing worth remembering is it is your image and your representation of the chosen target. As long as you are happy with the result that is realy all that matters. People like myself only give you an alternative representation of what we believe the target to look like, right or wrong we will never know but it is just another take on the same image.
The other reason for the rework or reprocess is an attempt to pass on methods we use to achieve our end results. Whether you use it or not is up to you as an individual. It is your image after all.
I have only put up an alternative to the original image. A few things worth remembering about astro imaging and in particular the QHY8 are The QHY8 has a sensitivity towards green, having twice as many green pixels in the Bayer grid as the other two colours. Ha is predominantly an orange red colour. The incidence of a reflection nebula and the presence of quite heavy dust clouds around Tarantula will in my opinion at least, tend to darken both the reflection nebula and the emmission nebula quite some.
The other thing worth remembering is it is your image and your representation of the chosen target. As long as you are happy with the result that is realy all that matters. People like myself only give you an alternative representation of what we believe the target to look like, right or wrong we will never know but it is just another take on the same image.
The other reason for the rework or reprocess is an attempt to pass on methods we use to achieve our end results. Whether you use it or not is up to you as an individual. It is your image after all.
Keep up the good work.
Totally agree with all that you say above Doug. I like your colours too. At the end of the day that's what the processing is about. Making a visually pleasing image. What I've always been wondering is what the "original colours" should look like. As Peter Ward posted once, we don't want to shoot in broad day light and found out that people have green skin tones. Then you know that the RGB ratios 1:1:1 are wrong. I wonder what ratios other QHY8 users use when doing RGB.
Regarding the QHY8 I know that it has twice as many green pixels as the two other channels. So what ratios would you use? Also have you found that shots around the Tarantula always seem to be noisier than other areas of the sky? Is this because there are so many stars in the area?
Totally agree with all that you say above Doug. I like your colours too. At the end of the day that's what the processing is about. Making a visually pleasing image. What I've always been wondering is what the "original colours" should look like. As Peter Ward posted once, we don't want to shoot in broad day light and found out that people have green skin tones. Then you know that the RGB ratios 1:1:1 are wrong. I wonder what ratios other QHY8 users use when doing RGB.
Regarding the QHY8 I know that it has twice as many green pixels as the two other channels. So what ratios would you use? Also have you found that shots around the Tarantula always seem to be noisier than other areas of the sky? Is this because there are so many stars in the area?
I Think the real reason for apparent noise when imaging in the Tarantula area is probably a combination of a couple of things.
1. The area is an area very rich in nebulosity and dust and we are probably, with long exposures picking up a lot of the fine dust areas which appear as noise.
2. The area being fairly close to the SCP makes guiding a little more dificult than other parts of the sky and we may well be suffering a little from image drift in some of these faint areas. You may or may not have noticed the dificulty gaining a calibration with programs such as PHD in this general area.
As for debayering ratio's, I use images plus and debayer at generic RGB and then do all manipulation in CS3 which allows me to manipulate colour and saturation to my own taste.
Colour balance and image saturation is an interesting topic which is dificult to really contemplate. The images we see are still entirely the final result of our own expectations. Some scientific basis must exist as to the colours used and final levels but I personally have no idea about this scientific data and generally develope an image to be pleasing to myself and usually based on other images of the same target.
WHO KNOWS. This would make an interesting discussion thread.
How did you combine the subs? SUM or MEAN? I found mean to keep star colors better and to be softer on the stars when stacking. Try different way of combining the data and it might soften the whole look.
Your advice is very useful, I've had a quick go at the original image and the improvement is amazing. 
