# 12 Feb, 2020 21:16
Hey there to all my fellow backyard astrophotography warriors.|
apologies in advance as I kinda know how most people answer my question, but I would really like to ask what kind of ratio other people use when combining Ha with RGB ie Ha total integrated exposure times against RGB total integrated exposure times.
I've looked at plenty of excellent images on Astrobin and the although combined exposure vary differently for obvious reasons, there does seem to be a general theme in that The combined Ha data normally greatly exceeds the Combined RGB data.
Reason I ask is that I’m currently working on my first attempt at a HaRGB image of the Flaming Star Nebula of which I’ve captured and stacked a solid 9 hours of Ha and could do with a little pointer on how much RGB would be a reasonable amount to integrate into the final image?
Finger in the air and a gut feel is I’m trying to capture at least the same amount of RGB as the Ha, but instinct is kinda telling me to go for more??
Any helpful advice or gentle pointers on what other folks have had good results from would be much appreciated.
Many thanks in advance, Stuart from a cloudy and stormy UK 👍🏻🧐🥶
# 13 Feb, 2020 04:28
Stuart. There is no cookbook recipe here. Experimentation would be the best way to establish your solution. The method I approach processing is very different from someone else. My choices may work for my data and system, but not for someone else. These systems are very different, and as a result, different approaches are taken. This is where you have to experiment and let that experience guide you. You will know that your data is lacking in one area, and enough in another. Repeating numbers verbatim generally is not a good solution as you want to understand the relationships of your data and what you are trying to create in the end.|
You can never get enough data. The more you have, the harder you can push it. Remember, Ha is very different to Lum and will not always support the under lying RGB component of your image. It will be too bright in some areas, and this will have an effect on your colour, sometimes washing it out. Every image I attempt is processed very differently from the last. It comes down to what relationships I can determine in the data before I start to process.
# 13 Feb, 2020 06:16
This is my opinion: Luminance give you details and contrast, RGB give you the colors. To have good details you need a lot of integration because you cannot exceed in denoising. For colors, you can push denoise far without problems, so even few shots can be enough.|
Remember: longer shots mean more signal, more shots mean less noise. Always take shots as long as you can (in ideal conditions I use 1800” for NB and for L, 900” for RGB).
Ha will be Luminance so 9hrs are a very good amount.
take in consideration gathering also L data: on the Flaming star (as in many nebulae) you also have a reflection part. If you are using just Ha, you will hide it. Spend some hour in L (or a light pollution filter) to get all information from the reflection nebula and from stars (Ha stars are very little and destroy stars in the HRGB image). Combine this L with Ha to get a Super-L to use on your RGB
# 13 Feb, 2020 09:11
Thank you very much for your input Terry. Yes, I kinda knew that was the answer I would receive. As with everything in this pastime, it is always a case of experimentation combined with trial and error. I couldn't agree more in that more data the better and the easier post processing will become with more integrated exposure. Originally my plan was to collect around 9 hours Ha and 18 hours RGB so around 27 hours total. My trouble is that I'm 3 months in now trying to collect the final RGB data with around 4 hours to date, but the remaining clear nights that I need are beyond elusive due to the blooming clouds and storms here in UK.|
Ill keep plugging on and hopefully get my remaining captures before my target gets to low in the sky.
Thank you for your valued feedback, much appreciated
# 13 Feb, 2020 09:30
In my (limited) personal experience using a one-shot colour camera (ASI294MC Pro), Ha used as Luminosity during processing can do wonders for the sharpness vs noise of your image for emission nebulae or for the emission part of a large nebula etc.! It also helps to reveal "hidden" or difficult parts of the nebula's surroundings, such as faint hydrogen gas.|
I haven't done HaRGB for galaxies yet, but I see results where Ha can add that extra punch in the red colours of the emission nebulae/starburst regions of a galaxy (you will not use Ha as Luminosity for galaxies though).
For emission nebulae, currently I am aiming to get the same amount of hours of exposure for Ha as for my RGB exposures, or even double it in some cases (i.e. for 5 hours of RGB, I aim to get around 5-10 hours of Ha). So, in the specific case you are asking about, I would try to add at least about 4-6 hours of good RGB data to get a good result.
I then combine Ha to RGB as luminosity using the "LRGB Combination" tool in PixInsight, where I usually set Luminosity (Ha) at 50-75%, testing which value gives me the best stars & star colours.
# 13 Feb, 2020 09:49
Thank you very much for your input on this. Yes my RGB will all be captured through my CLS filter. Having stacked my Ha data, the stack has come out very nice indeed. My next step is to process this in the Red channel in Photoshop and then use this as a luminance layer once I've collected and processed my RGB data.
Another quick question here and apologies up front as Im only 10 months into this hobby, but how do I create a separate luminance image from my RGB? It was my understanding that my Ha data would provide the luminance layer, but from what you've said, I can create another luminance image from my RGB data and combine this with my Ha and process together?
Just so I understand, is this a case of using my RGB data and process this in mono along with my Ha data? how would I achieve this please?
Thanks in advance
# 13 Feb, 2020 11:09
Thank you very much for your input on this.
My imaging camera is a modified Canon 200D DSLR, so can be rather noisy, as with most DSLR's, but plenty of combined exposure, darks & bias do a reasonable job of reducing this.
I hear what your saying in that Ha data you combine is usually around double your RGB, which makes perfect sense. TBH, I was concerned when I saw my first RGB subs come through as the signal on the nebula was very weak, but the stars and background look good. The issue I'm going to have I think is that my Ha data is very strong and I'm fearful that the stars in my Ha will wash out / destroy the colour in my RGB stars when I combine the images which Terry has pointed out above.
I think from what Im reading, I just capture as much RGB as possible and if the sky doesn't play ball, Ill aim for at least 5 hours of RGB as you have suggested and hopefully that will give me enough colour.
Really appreciate your feedback on this, thanks a bunch
# 13 Feb, 2020 12:14
Stuart Rawsonduplicate the image in PS and choose from the menu Image-Method-Grayscale
merge this L and the Ha in brighten mode
# 13 Feb, 2020 12:30
During my processing methodology (where I use Ha as L at 50-75% and combine with RGB), I have found that, since Ha generally has much smaller stars than a broadband (RGB) image, the outline of my stars will get tighter and double stars will become more resolved/apparent than in my RGB image. This is one of the reasons tracking problems are more unforgiving when imaging in narrowband (that, combined with the longer single exposures vs broadband). This is why I don't use the Ha/Luminance layer at 100%, but at 50-75% or whatever is best (based on my previous processing), so that the stars don't become white points with huge colourful halos. This might still happen, but if you are careful the effect will be negligible.
Another effect of using Ha as Luminosity is that parts of the emission nebula might become a lighter shade of red (closer to pink), which personally I don't mind, but you can try and (somewhat) remedy that with the colour balance or by recalibrating colour after the combination.
I am in no way an expert in this (and still have a LOOOONG way to go), as I started my astrophotography journey less than 1.5 year ago. But I find that, as others mentioned above, you can try and learn different processing techniques, but there is not one technique that works best for every set of data or for every target. I spend way too much of my free time trying different combinations of things before I am satisfied with an image.
# 15 Feb, 2020 13:35
Alessio ParianiThank you very much Alessio, really helpful instructions their. Much appreciated 👍🏻Stuart Rawsonduplicate the image in PS and choose from the menu Image-Method-Grayscale
# 15 Feb, 2020 13:38
Orestis PavlouStuart RawsonDuring my processing methodology (where I use Ha as L at 50-75% and combine with RGB), I have found that, since Ha generally has much smaller stars than a broadband (RGB) image, the outline of my stars will get tighter and double stars will become more resolved/apparent than in my RGB image. This is one of the reasons tracking problems are more unforgiving when imaging in narrowband (that, combined with the longer single exposures vs broadband). This is why I don't use the Ha/Luminance layer at 100%, but at 50-75% or whatever is best (based on my previous processing), so that the stars don't become white points with huge colourful halos. This might still happen, but if you are careful the effect will be negligible.
Thank you for your very informative reply Orestis. Really appreciate you passing some help my way and I’ll be giving your advice a run through once I’ve collected my RGB.
thank you very much
# 17 Feb, 2020 02:05
Stuart, this method by Marco Lorenzi is brilliant and works very well for me.|
# 17 Feb, 2020 08:55
|I think the Ha might be 2 times or more of RGB, it will get better resolution and contrast. Finally Light frame is very importont .have fun! Thank you very much!|
|You have no new notifications.|