Return to the Cloudy Nights Telescope Reviews home page


Astrophotography and Sketching >> CCD Imaging & Processing

Pages: 1 | 2 | 3 | (show all)
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: Do we need Luminance for imaging? or only RGB? new [Re: Ken Crawford]
      #5679346 - 02/14/13 12:20 PM

First, by the way, I modified my post above a bit, I realize really I meant to say "intensity" when I said frequency.

Anyways, just getting back to SNR, as I find this metric fascinating, I guess, I still haven't been convinced that SNR is really the best metric to describe image quality as we talk about it. For one, as mentioned in the exposure length thread, it ends up being a biased estimate, if skyglow goes up for example, it can actually improve on paper.


But also, regardless, pretending "observed SNR", did actually measure "true SNR" perfectly, the whole idea of talking about SNR as a metric, even under perfect conditions seems to fail somewhat. This is because when we do talk about 'image quality" we are talking about our perception of quality, not just means and standard deviations in the data. When we're talking about luminance, the means and standard deviations do seem to make more sense because of the way our eyes perceive grey scale and metrics used to describe luminance take into account the weighting of our eyes.

When we start talking about color though, this changes significantly because we are not considering the weight anymore. From a scientific perspective, it seems that we do want to know the means and standard deviations of the colors. However, our eyes can can be 20x or more, more sensitive to say green than red or blue. So, when we start talking about color, I think that there is a bigger discrepancy between the numbers and our perception. The luminance takes into account our perception, so, we can use that to boost our perception of the data and make a "better" image, although, as mentioned from a pure numbers perspective, it probably isn't lower variance, i.e. higher SNR.


So, when we start talking about color data, it seems we should almost use a luminance weighted estimate of the SNR. Does that make sense to others? And does anyone already use such an estimate?

Edit: also by the way, how do most programs measure SNR when using color data. For example, do they only look at variance of pixel to pixel intensity and ignore color? I think so, as when in Maxim DL, when I put the little curser over part of the image, it doesn't give me seperate L, R, G and B SNRs. So, when your talking about improvements to SNR etc... to lay ears such as mine, it may help to know the specifics of how it is being measured. If we are just measuring luminance SNR, then it seems it would improve with extra, separate L data. And if we're discussing SNR, with regards to what are we referring, just the measured value, or the theoretical, underlying value?

Edited by Inverted (02/14/13 12:52 PM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
freestar8n
Post Laureate
*****

Reged: 10/12/07

Re: Do we need Luminance for imaging? or only RGB? new [Re: Ken Crawford]
      #5679396 - 02/14/13 12:53 PM

Thanks Ken - That all looks good to me. I think it's great you were able to incorporate the color channels to improve the depth for the scientific goals of the project - and at the same time get nice color images of the galaxies to provide context for the star streams.

As for LRGB - I'm a believer that it *should* work to make a nicer looking result with more detail and better colors - but I think a key problem for me is slight spherochromatism in my c11 with reducer that makes it hard for the star sizes in different colors to match exactly so the end result looks natural. I assume I could mess with it in processing - but my main message is for people to give it a try, but it may not work as automagically as hoped - and there is no *inherent* improvement in the actual color SNR.

Thanks,
Frank


Post Extras: Print Post   Remind Me!   Notify Moderator  
Ken Crawford
sage


Reged: 06/02/09

Re: Do we need Luminance for imaging? or only RGB? new [Re: freestar8n]
      #5679469 - 02/14/13 01:24 PM

Quote:

Thanks Ken - That all looks good to me. I think it's great you were able to incorporate the color channels to improve the depth for the scientific goals of the project - and at the same time get nice color images of the galaxies to provide context for the star streams.

As for LRGB - I'm a believer that it *should* work to make a nicer looking result with more detail and better colors - but I think a key problem for me is slight spherochromatism in my c11 with reducer that makes it hard for the star sizes in different colors to match exactly so the end result looks natural. I assume I could mess with it in processing - but my main message is for people to give it a try, but it may not work as automagically as hoped - and there is no *inherent* improvement in the actual color SNR.

Thanks,
Frank




I even get some star size variances from seeing differences from night to night with my RC. I normally images several nights on a target. What I do is a very mild positive constraint Decon on the two largest FWHM color channels to push them down to the smallest of the color channel. I am talking very mild 10-20 iterations with a proper PSF. I then find my color fringing is less in the RGB master.

I also stretch around a star mask to keep the color from blowing out the Lum margins.

This is really good stuff Frank, your comments gives me inspriation for this hobby as I have not been as active as I would like to be on the imaging side. I learn from these types of interactions and gain more fire in the belly

Kindest Regards,


Post Extras: Print Post   Remind Me!   Notify Moderator  
alpal
Post Laureate


Reged: 06/15/09

Loc: Melbourne Australia.
Re: Do we need Luminance for imaging? or only RGB? new [Re: alpal]
      #5680271 - 02/14/13 09:49 PM Attachment (9 downloads)

Quote:

The Luminance has a higher signal to noise ratio than any particular colour of RGB.

Therefore in the faint areas of the picture -
luminance will enable us to see the low s/n data which would otherwise be invisible in the noise.

The only test would be on a very faint object such as a galaxy & in particular it's faint arms.




I tested the above theory out on some actual single frame RAW data from galaxy NGC 253.

I used a single Luminance frame & a Green frame.
Both frames were 3 minute exposures under light polluted skies with no ALP filter.
I cropped each of them out of the full size pic.
I then pasted them onto a new pic.
I stretched both of them equally as one pic.
I then stretched the Green version ( on the right hand side )
so that the background was equalised with the luminance frame. ( I chose 40 with the eyedropper in PS curves )
The area outlined as a box is then expanded later for the next pic to view right down at pixel level.


Post Extras: Print Post   Remind Me!   Notify Moderator  
alpal
Post Laureate


Reged: 06/15/09

Loc: Melbourne Australia.
Re: Do we need Luminance for imaging? or only RGB? new [Re: alpal]
      #5680276 - 02/14/13 09:51 PM Attachment (12 downloads)

Next is the expanded pixel version of just the box.
Notice the faint arms are hardly visible in the Green version on the right hand side.
The faint arms are lost inside the noise.


Post Extras: Print Post   Remind Me!   Notify Moderator  
blueman
Photon Catcher
*****

Reged: 07/20/07

Loc: California
Re: Do we need Luminance for imaging? or only RGB? new [Re: kfir Simon]
      #5680358 - 02/14/13 10:44 PM

I have not used true luminance for quite a while. I make sythetic luminance from the RGB. But I take a lot of RGB and then I have 30-60 frames to make a sythetic luminance. I have not found that luminace in needed for any of my images.
Blueman


Post Extras: Print Post   Remind Me!   Notify Moderator  
neptun2
scholastic sledgehammer


Reged: 03/04/07

Loc: Bulgaria
Re: Do we need Luminance for imaging? or only RGB? new [Re: blueman]
      #5680565 - 02/15/13 01:53 AM

Well if you create the luminance from the RGB this will not help in comparison with pure RGB. If you shoot separate luminance you get the whole wavelength spectrum. If you only take R, G and B you will miss some wavelengths because the filters' spectrums do not overlap completely. At least this is my understanding.

Post Extras: Print Post   Remind Me!   Notify Moderator  
freestar8n
Post Laureate
*****

Reged: 10/12/07

Re: Do we need Luminance for imaging? or only RGB? new [Re: alpal]
      #5680600 - 02/15/13 02:38 AM

Hi-

There is no question that L will probably have higher SNR than G for the same exposure time. But in LRGB, the claim is that multiplying G by L will increase the SNR of G - and that is why LRGB is good. That part is not correct because multiplying G by L doesn't improve the accuracy of G at all - in fact it corrupts it by mixing in R and B with the new G signal. But it ends up looking better in the end because you can really smooth the color channels before you do the LRGB combine - and your eye/brain won't realize it.

A big reason your G signal is fainter is that some of the structure in there looks like Ha stuff - in the red. It's supposed to be missing in G. If you mix L with G to make a 'better' G, you would be incorporating Ha into the G - but that makes no sense. Your G signal may look better and have less noise - but it's not accurately capturing G. So your signal is corrupted and your SNR for the G signal would go down - even if the end result looks better.

This is even more extreme when people do HaRGB. Some people like it, others don't - but you can take a nice, contrasty Ha image with a lot of detail - and multiply it by RGB and get an incredibly colored image. But did you improve the accuracy of the G signal? No - you just mixed G with Ha and got a colorful image. If you like the result that's fine - but it didn't improve the accuracy or SNR of the G signal - even though the colors look "better".

Frank


Post Extras: Print Post   Remind Me!   Notify Moderator  
alpal
Post Laureate


Reged: 06/15/09

Loc: Melbourne Australia.
Re: Do we need Luminance for imaging? or only RGB? new [Re: freestar8n]
      #5680605 - 02/15/13 02:42 AM

Hi Frank,
Yes I would probably have to repeat the test with Red to be sure.
Thanks for the insight.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: Do we need Luminance for imaging? or only RGB? new [Re: alpal]
      #5680772 - 02/15/13 07:25 AM

Hi Apal, also, I think you should also display as green, not grey scale. When you display it as green, the perceived difference will be different than in the precieved difference in grey scale. The luminance layer of an image is really adjusted to the sensitivity of our eyes. However, this will have the least impact for green, as our eyes are most sensitive to green. Well, actually to green/yellow, our visual perception peaks at about 550nm.

However, then, our eyes are not as sensitive to red and blue, so, if you take that green image and convert it from green to blue or red, I would expect the perceived "SNR" (or more properly stated in this case apparent detail) would be less.

This is an interesting result. It seems in these threads the focus is on SNR, as that is perceived as a quantifiable metric and "perception" of quality is apparently not. However, I'm not sure that is true. The characteristics of the human eye and perception of color are well defined by the photopic luminosity function. We can actually quantitatively determine color perception of an average human eye. And actually, when you convert color to a luminance layer, the software will generally weight accordingly, so, what you get in luminance is (assuming the software is weighting properly) is a representation of the perceived color, not a 1:1:1 scaled representation of actual color.

So, I realize when I asked about a color weighted SNR, a luminance layer actual is.

Anyways, that's just my .02. Years ago I worked in a molecular biology lab studying retinal rod and cone progenitor cells. So, this is one of the few topics in the hobby I actually should have good intuition about and perhaps even know something LOL.


Post Extras: Print Post   Remind Me!   Notify Moderator  
alpal
Post Laureate


Reged: 06/15/09

Loc: Melbourne Australia.
Re: Do we need Luminance for imaging? or only RGB? new [Re: Inverted]
      #5680861 - 02/15/13 08:42 AM Attachment (7 downloads)

Hi "Inverted",
Here is the same pic but green is green - what do you think?


Post Extras: Print Post   Remind Me!   Notify Moderator  
neptun2
scholastic sledgehammer


Reged: 03/04/07

Loc: Bulgaria
Re: Do we need Luminance for imaging? or only RGB? new [Re: alpal]
      #5680906 - 02/15/13 09:04 AM

At least my opinion is that there is still fainter visible detail in the luminance frame which is not present in the green channel.

Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: Do we need Luminance for imaging? or only RGB? new [Re: neptun2]
      #5681063 - 02/15/13 10:14 AM

Yes, the green I think is actually even a little less visible now than before when shown in grey (as would be expected). Now, if we took the green and converted it to red or blue, it should look different still. However, the kicker is, that in all instances, it's the same image and the SNR is the same. When we're looking at greyscale, we're just looking at contrast of the image, which our eyes see differently. It isn't effected by our perception of different colors, in the same ways and the SNR does not have a linear effect with respect to our perception of all colors.

When we take a color image though, the SNR is reduced because the filter is clipping some light. The SNR of the mono image is not effected in this way. Then we work on luminance and apply it to the color image. So, even though the SNR will not effect perception of colors linearly, increasing it (by utilizing more wavelengths as we are not clipping any with the filters) will still boost our perception of the colors.

I guess as Frank mentioned, you now may be seeing an increase in color because there are extra wavelengths being captured by the mono channel. I'm not entirely sure I'd agree this is "less accurate though" The mono "luminance" channel is the most accurate representation of the signal, as no signal is being clipped by the filter, however, they colors may not end up perfectly "correct". Although, I'd note, that when using filters, the filters we use aren't perfect, there is overlap and gaps etc.. so, the color also isn't "perfectly correct". With monochrome imaging though, with no filters, you are capturing more signal, so, the SNR should be greater for a given exposure and therefore there should be more contrast in the final image (even if the color isn't perfectly represented).


Post Extras: Print Post   Remind Me!   Notify Moderator  
JoseBorrero
Post Laureate
*****

Reged: 09/04/09

Loc: Puerto Rico Island
Re: Do we need Luminance for imaging? or only RGB? new [Re: Inverted]
      #5681249 - 02/15/13 11:44 AM Attachment (4 downloads)

At least on photoshop R,G,B can be layered and blend as "screen", then merge. isn't the same as Luminance concept?

by the other hand showing in this diagram that luminance fill some gaps between the colors, I might be a bad interpretation of a wrong color at all.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: Do we need Luminance for imaging? or only RGB? new [Re: JoseBorrero]
      #5681316 - 02/15/13 12:16 PM

I looked into how Photoshop processes luminance and it looks like if you go from an RGB image to a "luminance" image, it scales it as Luminance = 0.3 R + 0.59 G + 0.11 B. So, as expected, it's not a 1:1:1 ratio. It appears the values are an attempt to scale according to the luminosity function, based on the base values of the 3 colors as used.

Going the other way, I am not as sure about the scaling.

Also, I'm not quite following what you were saying about the graph?


Post Extras: Print Post   Remind Me!   Notify Moderator  
Peter in Reno
Postmaster
*****

Reged: 07/15/08

Loc: Reno, NV
Re: Do we need Luminance for imaging? or only RGB? new [Re: Inverted]
      #5681337 - 02/15/13 12:24 PM

PixInsight allows you to extract Luminance from RGB using 1:1:1 RGB ratio.

Peter


Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: Do we need Luminance for imaging? or only RGB? new [Re: Peter in Reno]
      #5681385 - 02/15/13 12:45 PM

Quote:

PixInsight allows you to extract Luminance from RGB using 1:1:1 RGB ratio.

Peter




but then it isn't "luminance" By definition, "luminance" is the weighted power intensity, weighted to the sensitivity of the human eye. The SI definition of luminosity is candella per square meter, where a candella is based on the photopic luminosity function as shown here :



Also, I think technically, we probably shouldn't call a monochrome image a "luminance image" until it has been processed and stretched to be better perceived by human vision. Hobby usage may vary though...

The processing steps we use then are sort of backwards, which I think is why it is perceived as "not as accurate". Although, I'm not sure color rendering in the hobby is done super accurately regardless. Perhaps somewhat if we do RGB and white balance, but it seems to really be accurate, the filters should match the response of the ccd, or be adjusted to it and not have much overlap, or gaps. I could be wrong, but I think scientific filters try to do this more thoroughly.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Peter in Reno
Postmaster
*****

Reged: 07/15/08

Loc: Reno, NV
Re: Do we need Luminance for imaging? or only RGB? new [Re: Inverted]
      #5681453 - 02/15/13 01:08 PM

It was suggested to me from PixInsight forum to use 1:1:1 RGB ratio if you want to create and process pseudo Luminance and combine with RGB. I find this method to work quite well and better than the ratio you listed.

It's mostly all about making the images look prettier. It's a personal preference. I live in pretty high LP area and I have to make some adjustments to get what I want to match as closely as other great images. Everyone's method is different based in their equipment and location. But if I live under a very dark sky like New Mexico, then processing RGB alone would be very easy without added Luminance. JWalk's examples are very good and he uses OSC camera. I have processed other people's images taken from dark skies and they were very easy to process because data is much better from dark skies and very little work to take care of unwanted signal (LP) or noise.

Either way, processing with pseudo or mono Luminance is not accurate in terms of color. Like Frank said, it allows you to make pretty pictures but not necessarily accurate colors.

Peter

Edited by Peter in Reno (02/15/13 01:16 PM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: Do we need Luminance for imaging? or only RGB? new [Re: Peter in Reno]
      #5681477 - 02/15/13 01:20 PM

For sure. I'm not talking about what works best, I'm just trying to take a step back and look at what we are actually talking about and how they relate to "pretty pictures".

Really, there isn't much green in space, so, weighting green 59% probably doesn't make much sense as it is probably mostly noise. So, using this function, we may be distorting our overall SNR unnecessarily for example. Speaking of PixInsight, I think there is a noise reduction function (I forget what it's called) that targets green specifically, as this is considered to likely be noise. The point though is metrics like SNR only work so well, and start breaking down when we move from quantitative data land to more qualitative human vision land...


Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: Do we need Luminance for imaging? or only RGB? new [Re: Inverted]
      #5681509 - 02/15/13 01:30 PM

Also, by the way, the monochrome CCD image I think is 1:1:1 until we stretch it. Once we stretch it, then it doesn't necessarily match the 1:1:1 RGB ratio anymore, but does provide the background for our perception of the RGB ratio (i.e. luminance). So, it is perhaps "less accurate". That's been mentioned by myself and others, previously, but I'm not sure how clearly...

Post Extras: Print Post   Remind Me!   Notify Moderator  
Pages: 1 | 2 | 3 | (show all)


Extra information
10 registered and 26 anonymous users are browsing this forum.

Moderator:  Oldfield, Scott in NC, bilgebay, WOBentley 

Print Thread

Forum Permissions
      You cannot start new topics
      You cannot reply to topics
      HTML is disabled
      UBBCode is enabled


Thread views: 2061

Jump to

CN Forums Home




Cloudy Nights Sponsor: Astronomics