by DavidLeodis » Wed Jan 30, 2013 11:39 am
stephen63 wrote:DavidLeodis wrote:Thanks for that link bystander, which is appreciated. My understanding, on reading through its information, is that the information brought up through the "luminance" link in the explanation is confusing as it does not seem to be giving the same meaning of luminance that luminance does in image data.
David. The advantage of using the luminance filter/data/channel is it's relatively high signal to noise ratio(SNR). The human eye perceives detail in the luminance image. Higher SNR means we can stretch the data more without the noise becoming objectionable. Your eyes don't detect noise as easily in the color channels, so a lower SNR is acceptable. If it does get excessively noisy, just back off of the stretch, reduce the saturation, or run a noise eduction action. I'm sure there are a lot of other ways to skin a cat, but that's the gist of it. Also, when adding the luminance on top of the RGB data, it doesn't get its own channel. There are only three, as far as I know! Hope this helps!
Thanks stephen63 for your help, which is appreciated.
What confused me about luminance in regard to the APOD is that the information brought up through the link is that luminance (putting it simply) is the brightness of an object, yet in the explanation it stated the "image is a composite with luminance taken from an image by the 8.2-m ground-based Subaru Telescope..." which does not seem to be using luminance as the brightness but rather as a filter.
[quote="stephen63"][quote="DavidLeodis"]Thanks for that link bystander, which is appreciated. My understanding, on reading through its information, is that the information brought up through the "luminance" link in the explanation is confusing as it does not seem to be giving the same meaning of luminance that luminance does in image data. :?[/quote]
David. The advantage of using the luminance filter/data/channel is it's relatively high signal to noise ratio(SNR). The human eye perceives detail in the luminance image. Higher SNR means we can stretch the data more without the noise becoming objectionable. Your eyes don't detect noise as easily in the color channels, so a lower SNR is acceptable. If it does get excessively noisy, just back off of the stretch, reduce the saturation, or run a noise eduction action. I'm sure there are a lot of other ways to skin a cat, but that's the gist of it. Also, when adding the luminance on top of the RGB data, it doesn't get its own channel. There are only three, as far as I know! Hope this helps![/quote]
Thanks stephen63 for your help, which is appreciated. :)
What confused me about luminance in regard to the APOD is that the information brought up through the link is that luminance (putting it simply) is the brightness of an object, yet in the explanation it stated the "image is a composite with luminance taken from an image by the 8.2-m ground-based Subaru Telescope..." which does not seem to be using luminance as the brightness but rather as a filter.