L-RGB with DSLR

[blinky]

Just pondering as you do when you cant do any imaging.... Anyway I wonder if DSLR images can be processed by setting the DSLR to B&W and taking a series of images and using these for a L channel? Do you think it would give any gain in image quality or would I be as well to convert the image to B&W in Photoshop and use that copy as a L channel?

[Psychobilly]

I use the second option... Make a copy of the OSC RGB image an convert it to B&W then copy this back into the RGB image as a separate layer and set its mode to luminosity... you can often be more aggressive with the sharpening and brightness/contrast curves etc with this L layer ... and vary the effects by altering the layers opacity...

I sometimes use multiple copies of the L and use different processing on each...

Have a play...

If ever the weather plays ball I am hoping to start using the 2" Ha filter to capture data to use for a HaRGB type image... or use the Ha as L...

Thinking about it a bit more you might be able to do something similar by just making a copy of the R channel and using that as L... hmmm time for a bit of playing I think...

Seeing as it is rather wet outside at the moment

Peter...

[Psychobilly]

Just had a quick play with some old NGC7000 IC5070 Data from the Equinoxe 66 and 350D...

I Seperated out the Red channel from the RGB used this Greyscale R data as a L/Ha channel with the original RGB but with more aggressive proccessing on it...

I quite like the result...

[ATTACH]24828[/ATTACH]

Anyway back to the plumbing now the last of the bits have arrived...

Peter...

[Psychobilly]

Just had a quick play with some old NGC7000 IC5070 Data from the Equinoxe 66 and 350D...

I Seperated out the Red channel from the RGB used this Greyscale R data as a L/Ha channel with the original RGB nut with more aggressive proccessign on it...

I quite like the result...

[brianb]

Anyway I wonder if DSLR images can be processed by setting the DSLR to B&W and taking a series of images and using these for a L channel? Do you think it would give any gain in image quality

No. DSLR images remain colour even when it's set to monochrome mode, in fact you can convert back from monochrome to colour with no loss whatsoever if you record in RAW.

I can't see any point in stripping down colour images taken with a Bayer filtered camera into RGB channels & using these to reconstruct an image except in the case where the channels are misaligned in the original image due to prismatic effects at low altitude when using long focal lengths.

[Milamber]

I was going to say that splitting into RGB and processing seperatley would be the only way but then thought Brian was right - especially since most processing packages allow you to adjust each colour seperatley anyway and you get to see the effect on the image instantly.

But then I thought that if you were to split the colours in you processing package, you could then use the RGB images to synthesize other layers, a Good Thing^TM in my eyes.

Arthur

[minimalist]

One of the techniques I use in my post processing doesn't even involve splitting luminance and rgb per se. In photoshop, you can create duplicate your original layer twice, and set the blend mode of one to Luminosity, and the other to Color. Uncheck the original layer and now you have essentially a luma layer and a color layer. Works very well imo, you can even manipulate the luma layer to enhance the brightness of one of the channels, like red (which I tend to overdo sometimes).

Daniel