Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

The Color in RGB


Recommended Posts

I've been  working my subs from M101 for the past two weeks every which way and trying to combine them into a passable freshman attempt at a CCD image but no matter what I do the color is drab at best and boosting saturation just looks horrible.  I am trying to wrap my head around where the color comes from and maybe someone can tell me if I'm on the right track.   The Red Green Blue channels only work if there is enough difference from channel to channel to allow those individual colors to show through other wise they cancel themselves out and you get grey?  I've seen images that have similar integration time to mine, 6x10minutes each in LRGB, and they have good color (much better than mine at least.)  That being said if you don't have enough data that differentiates the channels you will not get dramatic color from the combination of them?  Am I on the right track with this thinking?  Because if all I need is more data, I can do that. But if I am not processing them correctly then I need to start back at square one.  

#banging_head_against_keyboard

Link to comment
Share on other sites

Not sure I get your idea about color. The channels are supposed to balance. The "grey" would be your background. If the source of the light has more red, object will look red. If you attempt to unbalance the color channels, the background will have a color cast. If you boost the L, it will wash out the colors

I'm not an LRGB or NB imager. Experienced folks here have commented the NB is easier to process than LRGB. Here are some references:

http://www.astro-imaging.com/Tutorial/LRGB.html

http://starizona.com/acb/ccd/software/soft_total_ngc253pp.aspx

Post some finished work. Lot's of help here.--Jack

Link to comment
Share on other sites

if all else fails use multiple layers and colorize them, I know it might be cheating but you still took the image :)

Could,it actually be that easy, if so what do we spend a small fortune on LRGB filters ??

Would like to here opinions on this

Regards

AB

Link to comment
Share on other sites

IMO colorizing is fraud. Deal with the data. Don't invent it.

I agree, and I am not saying it is the way to go, but you are not inventing the data, that is what is captured, you are only putting colour to it....

This thread could get very interesting, I look forward to hearing comments in this, even though I already have a very good set of LRGB filters, the thing is I never even thought it could be done this way.

Regards

AB

Link to comment
Share on other sites

I only said that because we are doing this for a hobby, having fun.  why let it stress you over one image? if youve tried time and time again to "fix" it just leave it as is, or colour it!  there is always a next time and the conditions might be better then?  Dont stress yourself out, youve got a ton of money locked up in a hobby, that is pursued for enjoyment!! 

Link to comment
Share on other sites

Could,it actually be that easy, if so what do we spend a small fortune on LRGB filters ??

Would like to here opinions on this

Regards

AB

you spend the small fortune on these optics so you can capture the image how it naturally looks. and (even though i have no experience in it) once you nail that Nebula/planet/other perfectly using filters, i bet its much more rewarding

Link to comment
Share on other sites

I agree, and I am not saying it is the way to go, but you are not inventing the data, that is what is captured, you are only putting colour to it....

This thread could get very interesting, I look forward to hearing comments in this, even though I already have a very good set of LRGB filters, the thing is I never even thought it could be done this way.

Regards

AB

So color isn't part of data? News to me!

Link to comment
Share on other sites

So color isn't part of data? News to me!

No, because the filter in essence is just putting colour there too......

Taking a colour image through a say green, red or blue filter gives no real detail, the whole point of LRGB imaging is that all the detail comes from the L channel, and the RGB are there just to give colour, I watched the YouTube video that YOU linked too earlier in this thread and the guy in there said that. So it's only what I have been taught by that video,,as I had not much experience before that.

I assume you have watched the full half hour video?

Told you this would get interesting....lol

Regards

AB

Link to comment
Share on other sites

Each color channel has its differences.  Seemingly, where there are similarities that part of image seems to be grey. Where there are differences that is where I see color. I thought of this question when I got the the Green channel.  I did not refocus for any of the filters and the green seems bloated.  I learned to refocus each filter, but I went ahead and combined them anyways, I need the practice.  Each star has a green halo. This led me to the thought and the question.  Is a drab image caused by a lack of data or a problem in processing?

Link to comment
Share on other sites

The colourizing advice is, to be honest, way off the mark and would lead imaging into pure invention, which it isn't. You could do that if you just shot luminance.

To get the best out of your genuine colour data you have a number of techniques available and boosting saturation should be the last resort because it is also a great booster of noise. Briefly I'd do something like this:

Process and stretch the RGB. Process for low resolution (so don't do any sharpening) and low noise. Noise reduction can be quite nasty but on RGB it can be done without too much damage. Convert the RGB to Lab Colour Mode in Ps and then split the channels and use the bog standard Brightness and Contrast tool on the a and b channels. Increase the contrast by a whacky amount like 35 and recombine the channels. Convert back to RGB mode.

Make 2 copy layers. Set the blend mode on the top layer to Blend Mode Soft Light and flatten onto the second layer. Set the second layer to Blend Mode Colour, give it a Gaussian Blur of about 0.6 and flatten. If colour is still weak, repeat.

Now process your luminance for high contrast and sharp detail (but avoid sharpening stars.)

Paste your L onto your RGB in Blend Mode Luminance. Is the L washing out the colour? If so set its opacity to 30%, boost the saturation in the RGB layer, blur the L by 0.6 and flatten. Re-apply the L and repeat the whole process. When you make the final application of L at full opacity don't apply the blur. You will get all the sharpness of the original L.

Nobody said it was easy!!!  :grin:

Olly

Link to comment
Share on other sites

This is the way I combine process images except I don't use masks, I simply use different layers and work on each layer individually also using some of Olly's tips.  

One thing he didn't mention in the video was balancing the colours.

Neither did he mention registering the different filters so that they will sit on top of each other precisely.

Carole 

Link to comment
Share on other sites

The colourizing advice is, to be honest, way off the mark and would lead imaging into pure invention, which it isn't. You could do that if you just shot luminance.

To get the best out of your genuine colour data you have a number of techniques available and boosting saturation should be the last resort because it is also a great booster of noise. Briefly I'd do something like this:

Process and stretch the RGB. Process for low resolution (so don't do any sharpening) and low noise. Noise reduction can be quite nasty but on RGB it can be done without too much damage. Convert the RGB to Lab Colour Mode in Ps and then split the channels and use the bog standard Brightness and Contrast tool on the a and b channels. Increase the contrast by a whacky amount like 35 and recombine the channels. Convert back to RGB mode.

Make 2 copy layers. Set the blend mode on the top layer to Blend Mode Soft Light and flatten onto the second layer. Set the second layer to Blend Mode Colour, give it a Gaussian Blur of about 0.6 and flatten. If colour is still weak, repeat.

Now process your luminance for high contrast and sharp detail (but avoid sharpening stars.)

Paste your L onto your RGB in Blend Mode Luminance. Is the L washing out the colour? If so set its opacity to 30%, boost the saturation in the RGB layer, blur the L by 0.6 and flatten. Re-apply the L and repeat the whole process. When you make the final application of L at full opacity don't apply the blur. You will get all the sharpness of the original L.

Nobody said it was easy!!!  :grin:

that is the best troubleshoot/tutorial on image processing ive seen!

Link to comment
Share on other sites

  • 4 weeks later...

Each color channel has its differences.  Seemingly, where there are similarities that part of image seems to be grey. Where there are differences that is where I see color. I thought of this question when I got the the Green channel.  I did not refocus for any of the filters and the green seems bloated.  I learned to refocus each filter, but I went ahead and combined them anyways, I need the practice.  Each star has a green halo. This led me to the thought and the question.  Is a drab image caused by a lack of data or a problem in processing?

I see that you are using Astronomik filters from you signature. I had that same problem of star bloat and halos. It may be caused by your filters. Astronomik are now producing a new set of filters with better coatings. At present they only have the narrow band set done. But at some time in, hopefully the near future, they will produce the LRGB set with the same coatings.

Now for the important bit !!

Astronomik give a guarantee that they will replace the filters if found to be producing halos. There are instructions how to go about claiming on their website. You will need receipts from the supplier, your affected images and one or two other things about the image train.

I have just received my replacements free of charge, thus proving the do honour their guarantee.

Have a look at

Supplier Reviews here on SGL for my review.

Derek

Link to comment
Share on other sites

I see that you are using Astronomik filters from you signature. I had that same problem of star bloat and halos. It may be caused by your filters. Astronomik are now producing a new set of filters with better coatings. At present they only have the narrow band set done. But at some time in, hopefully the near future, they will produce the LRGB set with the same coatings.

Now for the important bit !!

Astronomik give a guarantee that they will replace the filters if found to be producing halos. There are instructions how to go about claiming on their website. You will need receipts from the supplier, your affected images and one or two other things about the image train.

I have just received my replacements free of charge, thus proving the do honour their guarantee.

Have a look at

Supplier Reviews here on SGL for my review.

Derek

Thanks for the info, Derek. I will check out their site on the halo issue. I have the LRGB IIc which I thought sorted the halos and I got their 12nm NB. The Astronomik are quite a bit less expensive here in the states which is why I had to get them over the Baaders. I spent my budget on the CCD.

I think I answered my original question when I persisted with M101 and got three more nights on it for over 18 hours of light. I spent an lot more time learning some processing skills and came up with this:

gallery_36993_3705_5424641.png

Sent from my SAMSUNG-SM-G870A using Tapatalk

Link to comment
Share on other sites

I'm finding this thread very interesting despite being a DSLR imager rather than CCD.

Having just watched the very interesting video linked above on how to combine LRGB data sets in Photoshop ....

What does adding the luminance data in a luminance layer do to the RGB layer?

Sorry this is slightly off topic.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.