H alpha and 'L' Luminance files

[tgervais]

I have been going through Harry's tutorial and looking at other postings as well - and Luminance files are mentioned a lot.
I am not sure what is meant when I am told to use my Luminance file.  Or my H alpha file. 

I am not completely lost though  because I think it is the 'gray' file that I get when I break down the colors .  I get a gray file and the R G B files.  Is the grey file the luminance file?

If someone could explain how to get a Luimance layer or / and a H alpha  file   it would be appreciated.

[aworonow]

Very briefly. Others can say more. If you take images through red, green, and blue filters, you end up with gray-scale images for each of those that can be calibrated and stacked separately into R, G, B integrated images. Using a color-combining procedure, those 3 images can be combined to create an "RGB color image." And that is all you really need, But, if you also took images with images through no color filter at all, that would be an L image (also gray-scale). The procedure LRGB Combine could augment the RGB image with the L image. (This is done once the images are nonlinear.) Again, this is all you would need for your image, but if you also took images through an hydrogen-alpha filter, then, using the script NBRGB, you could combine it with the RGB--really with the red component. This is best done before going nonlinear and before introducing the L into the mix.

This assumes your are using a mono camera, not a one-shot color (OSC) or DSLR camera. They produce the RGB simultaneously, but you can still add the Ha, however, there's really little point (as far as I know) of forcing such a camera to take L images, which are optional even in the situation described above. L images have often been used to bolster exposure detail and cut down on the total exposure time by taking fewer R,B,G images.

Alex

[Geoff]

With a monochrome camera it's usual to use 4 filters when assemblng a "normal colour" picture, ie not a narrowband one. These are red, green, blue and luminance. The luminance filter lets in the full spectrum, but usually blocks out the infrared radiation. The readon for putting a clear filter, rather than an empty slot is so that we don't have to change focus. The luminance image will have a higher SNR than the colour filters. The basic idea behind using a luminance filter is that the eye is more sensitive to brightness detail rather than colour detail, so that by combining a high SNR luminance image with a colour image assembled from the RGB filters you end up with a better looking final image.

You can also use other filters which let through a very limited range of wavelengths. These are the narrowband filters, most commonly hydrogen alpha,  oxygen III and sulphur II.

[tgervais]

Well thanks guys for the help and thoughts on this subject.

My problem is that I only have a DSLR camera which is what I am taking images with.   I have no filters etc..
And it is with these images that I am trying to use with Harry's Tutorials.  Maybe that is not a good thing.  But I thought if I could apply his lessons to what I have on hand,  than it means more.

So when Harry says to take your 'L' Luminance image I just simply use pixinsight to break down the color image I have into r g b and grey.
I follow the lesson using those files. 

I guess that may not be the right thing.  And probably will mislead me as the lessons become more difficult.  But at least the input that you guys provided tells me that I need to consider the proper layers and not to think about using a grey layer as I am doing.

Thanks again guys. 

[pfile]

if you want to make a proper L file from your RGB master, make sure to use the RGBWorkingSpace process to change your gamma to 1.0 and the luminance coefficients to 1,1,1 on the RGB image (apply the RGBWorkingSpace process to the RGB image), then extract L*. otherwise the L* you extract will be more heavily biased toward the green channel. you should then undo the RGBWorkingSpace process on the RGB image before you continue working on it.

another strategy would be to extract the R/G/B channels from the RGB master, save them, and then use ImageIntegration to create the L master by integrating all 3 of those channel masters. this is probably a better way to do it since by default the weighting of the channel masters will be by SNR.

rob

[tgervais]

if you want to make a proper L file from your RGB master, make sure to use the RGBWorkingSpace process to change your gamma to 1.0 and the luminance coefficients to 1,1,1 on the RGB image (apply the RGBWorkingSpace process to the RGB image), then extract L*. otherwise the L* you extract will be more heavily biased toward the green channel. you should then undo the RGBWorkingSpace process on the RGB image before you continue working on it.

another strategy would be to extract the R/G/B channels from the RGB master, save them, and then use ImageIntegration to create the L master by integrating all 3 of those channel masters. this is probably a better way to do it since by default the weighting of the channel masters will be by SNR.

rob

Thanks Rob.  I think that might work.  I need something to work with while I follow the tutorials and this could be the way.

Thanks again for your input.  It really helps at this point in my learning of Pix Insight.