How do I get a colour image with my RGBL data?

[Chris]

HI,

I took my first steps in RGB imaging using my new Brightstar Mammut camera last night and I'm not having much luck finding tutorials or info on how to combine it to form a colour image?

I've shot 60 minutes luminence and 40 minutes per RGB channel all binned 1x1. Due to the large 8.6 um pixel size of my camera I thought it would be a bad idea to bin 2x2 for the RGB.

Software wise I use NIghtwatch for capture, DSS for stacking, and Paint Shop Pro or a basic version of PS for processing.

If anyone can help with advice or links on how to get a colour image from my data I would really appreciate it, I'm a bit stumped! All I know so far is that I need to process each channel seperately and somehow assign colour to each.

Cheers

Chris

[swag72]

You stack all of your RGB data separately so that you 3 images. You then need to align them so that the stars are all in the same place in the frame. I don't know of any free software that does this, I use PixInsight and I know that many people use Registar. From there you go into photoshop and add each channel stack inro the relevant channels. There's a channel tab in the later versions of CS (don't know about your's sorry) and you basically copy and paste the red into the red channel and so on.

I saw this earlier but didn't answer as I didn't know how useful my reply would be as I don't know how you can align each image and even if your processing programme has a channels function. Hopefully that's given you the general idea. You end up with an RGB image, all aligned and with the channels matched as they should. From there .... good luck!!

[AndyUK]

Hi Chris - You can use DSS to align your separate R, G and B channels .

If you stack the first channel (say red), and make a note of the best frame (the one at the top of the list, which DSS will have assigned as the reference frame) when you come to stack the other channels you can put that red reference file in group one and force DSS to make it the reference frame. The only thing you have to remember is to uncheck it after aligning and computing offsets, otherwise it will be included within the stack...!

[Chris]

You stack all of your RGB data separately so that you 3 images. You then need to align them so that the stars are all in the same place in the frame. I don't know of any free software that does this, I use PixInsight and I know that many people use Registar. From there you go into photoshop and add each channel stack inro the relevant channels. There's a channel tab in the later versions of CS (don't know about your's sorry) and you basically copy and paste the red into the red channel and so on.

I saw this earlier but didn't answer as I didn't know how useful my reply would be as I don't know how you can align each image and even if your processing programme has a channels function. Hopefully that's given you the general idea. You end up with an RGB image, all aligned and with the channels matched as they should. From there .... good luck!!

Well I really appreciate you returning to the thread to reply Swag as I now have a general idea at least. Maybe the Nightwatch software which comes with the camera allows for alligning the stars in each of the stacked and processed channels so I should investigate this. I do have a basic PS so I'll see if it has a channel tab:)

Many thanks

Chris

p.s. Congrats for getting your hands on the legendary Tak!

[Chris]

Hi Chris - You can use DSS to align your separate R, G and B channels .

If you stack the first channel (say red), and make a note of the best frame (the one at the top of the list, which DSS will have assigned as the reference frame) when you come to stack the other channels you can put that red reference file in group one and force DSS to make it the reference frame. The only thing you have to remember is to uncheck it after aligning and computing offsets, otherwise it will be included within the stack...!

Thats a great idea Andy I'll give it a go if the camera software doesn't do it. Might it be worth me getting a copy of pixinsight? I've heard good things about it although I need to check out the price.

Thank you

Chris

[AndyUK]

Hi Chris - I have PS CS5, and therefore I can't really advise on Pixinsight. I have considered it a few times, and there are some VERY useful features in there, but there is a fair amount of crossover and personally I can't justify the additional expense of €171 (assuming you're not from a Eurozone country in which case it's €207).

However, I would suggest that PS CS2 will do most of what you really need to do (levels, curves, channels, various filters) and of course there are also various plug-ins you can get as well (HLVG (free!), Noels / Annie's actions, Gradient Xterminator etc etc), and the internet is awash with PS astro tutorials...

Having said that though, (and not wanting this thread to descend into a PS vs Pixinsight war ), you can get a 30 day trial of Pixinsight and Harry Page has some excellent tutorials on how to use it... In the right hands, it can certainly produce some beautiful images

[earth titan]

I've just gone through the same learning curve so here is what I did:

In DSS

• Stack the luminance frames in DSS with the relevant calibration frames
  
• Remove all but one of the luminance frames from the list and ensure it is not ticked
  
• Right mouse click on the remaining luminance frame and set as reference frame. This aligns the colours with the luminance frames
  
• Drag in and then stack the RG and then B in turn, using relevant calibration frames and keeping the same luminance reference frame
  

You now have four images - one of each LRGB

In PS

• Stretch the luminance layer and save it. This is easiest if you set the image mode to RGB (thanks Danny)
  
• In turn, open the colour frames, resize to match the luminance and keep them open as separate images
  
• On the channels tab (next to layers) pick 'merge channels'
  
• From the options pick RGB and when asked tell PS which frame is which
  
• Hey presto - a colour Image
  
• This can be stretched enhanced etc.
  
• Now drag the luminaince frame onto the RGB and position. Dropping the opacity of the luminince layer helps to see the layers.
  
• Once happy check the blending mode to 'luminiance' and hey presto.
  

Sure I've missed loads and others will correct me, but it seems to work as a start. Loads to learn though. Stick at it.