Ad
Ad
Ad
Pages: « 1 2 3 [4] 5 »   Bottom of Page
Print
Author Topic: Increasing color saturation in LAB: S-curve or straight line?  (Read 13597 times)
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6934


WWW
« Reply #60 on: April 30, 2013, 10:54:53 AM »
ReplyReply

Combinations of Saturation, Vibrance, HSL and Calibration Panel sliders in ACR/LR can do the same Sat/Lum disconnect as in Lab.
.................. No need to go into Lab in Photoshop to apply something as simple as increased saturation. What a cumbersome workflow to have to go into Lab just to add saturation.


Exactly right, and this has been the case since years back.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
MarkM
Sr. Member
****
Offline Offline

Posts: 327



WWW
« Reply #61 on: April 30, 2013, 11:21:40 AM »
ReplyReply

I don't know what you mean by "perceptually uniform".

It means that moving moving similar distances within the color space will result in similar perceptions of color difference regardless of where in the color space you are. Or to put another way the distance in the color space is uniformly correlated with the magnitude of the color difference.
Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6934


WWW
« Reply #62 on: April 30, 2013, 12:20:50 PM »
ReplyReply

OK, so this goes back to the CIE measurement experience deployed back in the 1920s to delineate and characterize human visual perception. A whole subject on its own, and probably fundamental to much of the colour work we do today, but its specific relevance to the use of the Lab space vs RGB for saturation editing in Photoshop remains somewhat elusive to me.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
MarkM
Sr. Member
****
Offline Offline

Posts: 327



WWW
« Reply #63 on: April 30, 2013, 01:55:50 PM »
ReplyReply

Quote
Mixing white light with a monochromatic light does not produce a constant hue, but Lab assumes it does! This is seen in Lab modelling of blues. It's the cause of the dreaded blue-magenta color issues or shifts. Lab is no better, and in many cases can be worse than a colorimetrically defined color space based on real or imaginary primaries.

Interesting. Never heard of monochromatic light, Andrew. What is it and how does it change hue perception that's different from "White Light" Lab doesn't take into account? How does that fit into Bruce Lindbloom's "Blue turns purple" analysis posted earlier? Or are you referring to something different from Bruce's explanation.

Monochromatic light is light of one wavelength, i.e spectral colors.

I think Bruce and Andrew are saying the same thing. It's very easy to see the problem if you plot your working or monitor's color space in LAB. Do it and look at the blue primary. Imagine drawing a vertical line (parallel to the L* axis) through that blue point. As you move up that line you move out of the gamut of your monitor. No look over that closest point on the surface of your monitor's space. What hue do you see? Purple right? If you take the route from the out-of-gamut LAB color directly to face you end up with hue shifts. I've attached a rudimentary (eye-balled) graph.

I also attached a TIF in LAB space that illustrates the problem. In the file the a* and b* numbers are constant. The only thing that changes is L*. The hue change is obvious. It's very easy to think you are simply changing luminosity when working in LAB but you're often unintentionally moving out of gamut and changing hue. You can add a Hue/Saturation adjustment layer to this image and move the hue slider to this and see how different colors exhibit this problem in varying degrees.
« Last Edit: April 30, 2013, 02:49:35 PM by MarkM » Logged

red2
Newbie
*
Offline Offline

Posts: 18


WWW
« Reply #64 on: April 30, 2013, 02:03:51 PM »
ReplyReply

One can test his/her browser by viewing an image in the ICC site. I viewed this site using the latest versions of Chrome, Firefox, Internet Explorer, and Safari. Chrome is partially color managed and the other browsers now appear fully managed as judged by this test. With wide availability of color managed browsers, there is little need to post images in sRGB if your audience is at least slightly sophisticated in color management (which I think would apply to most members of this forum).

Regards,

Bill


FWIW – I viewed the linked image in Safari v5.1.9 and Firefox v19.0.2 and 20.0. The Safari image looked like what is illustrated for a system supporting ICC version 4 and 2 profiles. The Firefox image looked like what is illustrated for a system supporting ICC version 2 profiles only. This was on a Mac running OS X 10.6.8 using a NEC PA271W monitor. I didn't try Chrome or any other browser.
Logged

Regards,
Bob D.
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #65 on: April 30, 2013, 02:46:22 PM »
ReplyReply

No, it is not just a matter of personal preference - it is technical. ANY departure from linearity will impact colour balance because you are altering the relationship between green and magenta on the a curve and blue and yellow on the b curve at different luminosity levels depending on where the non-linearity sets in, and if those linear curves do not intersect in the middle the colour balance of the whole image is thrown-off. Lab conversion is a needless and cumbersome complication from the get-go for 99% of the image editing the great majority of us ever need to do. There are many things in Photoshop that don't work in Lab mode, so most often you will need to convert back to RGB, and once you do this your Lab adjustments are baked-in and non-reversible, unless you have a separate duplicate image copy with the Lab adjustments converted back to RGB but layered-in. Lab has certain specialized uses that justify its inclusion in the Photoshop arsenal, but making simple adjustments to saturation is not one of them when there are much more straightforward ways of doing this in RGB.
This is not correct. You're thinking about how RGB curves work. In LAB mode, the the 'a' curve does not have a luminosity component, it only adjusts red/green saturation. A symmetrical 'S' curve that isn't so extreme as to have inversions will not throw off the color balance in the way you're suggesting. It will just increase saturation of the less-saturated colors affected by the steep section of the curve more than already-saturated colors that fall along the flatter section of the curve.

Having said that, I think an easier to way to boost just the less-saturated colors is to bring the end-points in, and then adjust the blend-if sliders of the layer properties to feather out the effect on the more saturated colors.

Quote
Lab conversion is a needless and cumbersome complication from the get-go for 99% of the image editing the great majority of us ever need to do.
Agree to disagree, I find it sometimes useful and just because you don't use it doesn't mean nobody should be using it.
Logged

Schewe
Sr. Member
****
Offline Offline

Posts: 5475


WWW
« Reply #66 on: April 30, 2013, 02:50:16 PM »
ReplyReply

A whole subject on its own, and probably fundamental to much of the colour work we do today, but its specific relevance to the use of the Lab space vs RGB for saturation editing in Photoshop remains somewhat elusive to me.

RGB is essentially based on our eye's ability to see trichromatic color. L*a*b* is a different color theory based on opponency of our eye's receptors. In fact, both theories are correct as our eyes have both trichromatic and opponency aspects.

In terms of using Lab for color correction, there's a school of thought that adjusting the tone curve of the lightness is better then adjusting the curves in RGB because with lightness, the color component is completely removed. There's a fellow with the initials DM that claims that adjusting curves in RGB destroys color hue and saturation. Another fellow whose initials are TK feels adjusting contrast in curves SHOULD modify the saturation (while trying to preserve the hue accuracy).

In Photoshop you can do either...and you don't need to convert to Lab to do so. You can simply set the blend mode of a curves layer to luminance–which while not EXACTLY the same as lightness in Lab, is close enough for the purposes of adjusting the tone curve without effecting hue and sat.

DM think's TK is wrong and as a result doesn't think Camera Raw nor Lightroom can be used for "professional" color correction. On the other hand, TK kinda started this whole digital imaging thingie...and DM is a "grumpy old white guy" still trying to be relevant.

L*a*b* can be useful...but I can't remember that last time I had to resort to converting to Lab in Photoshop for any corrections I could only do in Lab.
« Last Edit: April 30, 2013, 02:52:04 PM by Schewe » Logged
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #67 on: April 30, 2013, 02:55:28 PM »
ReplyReply

Combinations of Saturation, Vibrance, HSL and Calibration Panel sliders in ACR/LR can do the same Sat/Lum disconnect as in Lab.
Gee that doesn't sound "cumbersome" at all.  Roll Eyes

Quote
No need to go into Lab in Photoshop to apply something as simple as increased saturation. What a cumbersome workflow to have to go into Lab just to add saturation.
It's not just a simple saturation bump. A/B curves allow you to increase both saturation and separation of colors that need it without over-saturating colors that don't, and without screwing up the luminosity/contrast of the image. It works quite well for images that need it and is easier and more effective than trying to achieve the same effect in RGB mode.

Quote
I couldn't imagine having to do that on my 3000 Raw shots.
I can't imagine having to do anything on 3000 Raw shots. You and I are obviously different types of photographers. Doesn't mean I'm doing anything wrong just because I spend more time on individual images than you do.
Logged

Tim Lookingbill
Sr. Member
****
Offline Offline

Posts: 1154



WWW
« Reply #68 on: April 30, 2013, 02:58:08 PM »
ReplyReply

Quote
Now look over that closest point on the surface of your monitor's space. What hue do you see? Purple right?

Excellently explained, MarkM.

So this "Blue To Purple" shift in the color managed display preview has nothing to do with the gamut size of the display. I tended to think my sRGB-ish small gamut display was amplifying this effect but seeing others with wider gamut display see the same thing corrected that assumption.

All those blended color curves represented in that 3D color model actually shows how mapping errors occur using a straight line A-to-B approach converting from one color space to another. Looks to me like a lack of precision in plotting just where these primaries blend together to form other colors within each 3D model representation of a color space.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6934


WWW
« Reply #69 on: April 30, 2013, 03:00:41 PM »
ReplyReply

This is not correct. You're thinking about how RGB curves work. In LAB mode, the the 'a' curve does not have a luminosity component, it only adjusts red/green saturation. A symmetrical 'S' curve that isn't so extreme as to have inversions will not throw off the color balance in the way you're suggesting. It will just increase saturation of the less-saturated colors affected by the steep section of the curve more than already-saturated colors that fall along the flatter section of the curve.

Having said that, I think an easier to way to boost just the less-saturated colors is to bring the end-points in, and then adjust the blend-if sliders of the layer properties to feather out the effect on the more saturated colors.
Agree to disagree, I find it sometimes useful and just because you don't use it doesn't mean nobody should be using it.


Yes correct - after I wrote that I realized I should have been talking saturation, not luminosity.

And I didn't say it wasn't ever useful. It can be. It's just that most of the time there are easier ways.............I think there is a broad level of experienced support to this perspective.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6934


WWW
« Reply #70 on: April 30, 2013, 03:02:44 PM »
ReplyReply

RGB is essentially based on our eye's ability to see trichromatic color. L*a*b* is a different color theory based on opponency of our eye's receptors. In fact, both theories are correct as our eyes have both trichromatic and opponency aspects.

In terms of using Lab for color correction, there's a school of thought that adjusting the tone curve of the lightness is better then adjusting the curves in RGB because with lightness, the color component is completely removed. There's a fellow with the initials DM that claims that adjusting curves in RGB destroys color hue and saturation. Another fellow whose initials are TK feels adjusting contrast in curves SHOULD modify the saturation (while trying to preserve the hue accuracy).

In Photoshop you can do either...and you don't need to convert to Lab to do so. You can simply set the blend mode of a curves layer to luminance–which while not EXACTLY the same as lightness in Lab, is close enough for the purposes of adjusting the tone curve without effecting hue and sat.

DM think's TK is wrong and as a result doesn't think Camera Raw nor Lightroom can be used for "professional" color correction. On the other hand, TK kinda started this whole digital imaging thingie...and DM is a "grumpy old white guy" still trying to be relevant.

L*a*b* can be useful...but I can't remember that last time I had to resort to converting to Lab in Photoshop for any corrections I could only do in Lab.

That was the whole discussion in a nutshell. Very well summarized.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
bjanes
Sr. Member
****
Offline Offline

Posts: 2794



« Reply #71 on: April 30, 2013, 03:41:55 PM »
ReplyReply

FWIW – I viewed the linked image in Safari v5.1.9 and Firefox v19.0.2 and 20.0. The Safari image looked like what is illustrated for a system supporting ICC version 4 and 2 profiles. The Firefox image looked like what is illustrated for a system supporting ICC version 2 profiles only. This was on a Mac running OS X 10.6.8 using a NEC PA271W monitor. I didn't try Chrome or any other browser.


That is interesting. I was using Windows8 and a NEC wide gamut monitor calibrated with Spectraview.

Bill
Logged
jrp
Jr. Member
**
Offline Offline

Posts: 77


« Reply #72 on: April 30, 2013, 05:58:51 PM »
ReplyReply

This is not correct. You're thinking about how RGB curves work. In LAB mode, the the 'a' curve does not have a luminosity component, it only adjusts red/green saturation.

This is right in principle, but not so in practice.  When photoshop finds super light or super dark (saturated) colors it tries to make them displayable, which can affect the luminosity.

For my part, I like to be able to adjust luminosity / contrast separately from color as controlling one "variable" at a time is easier to my mind.  That can, of course, be done as Schewe suggest, in RGB, but this curve steepening technique can produce more pleasing results (probably because it separates the colors more, as someone else has pointed out).
Logged
red2
Newbie
*
Offline Offline

Posts: 18


WWW
« Reply #73 on: April 30, 2013, 09:37:56 PM »
ReplyReply


That is interesting. I was using Windows8 and a NEC wide gamut monitor calibrated with Spectraview.

Bill

My monitor was also calibrated (using Spectraview II).
Logged

Regards,
Bob D.
Gulag
Full Member
***
Offline Offline

Posts: 215


« Reply #74 on: April 30, 2013, 10:33:08 PM »
ReplyReply

First of all, don't sweat the small stuff. Conversions between different color modes won't destroy your images if you know what you're doing. As there are more than fifty ways to leave your lover, there are more than fifty ways to fiddle the curves in LAB mode. Want to learn more. Read this:

The Jacob’s Ladder Color Correction Technique



also check out this Yahoo discussion thread
« Last Edit: April 30, 2013, 10:37:06 PM by Gulag » Logged

“For art to be art it has to cure.”  - Alejandro Jodorowsky
digitaldog
Sr. Member
****
Offline Offline

Posts: 9100



WWW
« Reply #75 on: May 01, 2013, 09:19:31 AM »
ReplyReply

First of all, don't sweat the small stuff. Conversions between different color modes won't destroy your images if you know what you're doing.

It sure could if you're silly enough to do this in 8-bit per color!

A keeper on the sillyness of conversions not needed by the late great Bruce Fraser in response to the Lab conversion craze started years ago. What Bruce wrote then is as true  today. Plus I don't see anything in your link that couldn’t be achieved faster, with no data loss in a good raw converter.

Quote

Let me make it clear that I'm not adamantly opposed to Lab workflows. If
they work for you, that's great, and you should continue to use them.


My concern is that Lab has been oversold, and that naive users attribute to
it an objective correctness that it does not deserve.


Even if we discount the issue of quantization errors going from device space
to Lab and vice versa, which could be solved by capturing some larger number
of bits than we commonly do now, (though probably more than 48 bits would be
required), it's important to realise that CIE colorimetry in general, and
Lab in particular, have significant limitations as tools for managing color
appearance, particularly in complex situations like photographic imagery.


CIE colorimetry is a reliable tool for predicting whether two given solid
colors will match when viewed in very precisely defined conditions. It is
not, and was never intended to be, a tool for predicting how those two
colors will actually appear to the observer. Rather, the express design goal
for CIELab was to provide a color space for the specification of color
differences. Anyone who has really compared color appearances under
controlled viewing conditions with delta-e values will tell you that it
works better in some areas of hue space than others.

For archival work, you will always want to preserve the original capture
data, along with the best definition you can muster of the space of the
device that did the capturing. Saving the data as Lab will inevitably
degrade it with any capture device that is currently available. For some
applications, the advantages of working in Lab, with or without an LCH
interface, will outweigh the disadvantages, but for a great many
applications, they will not. Any time you attempt to render image data on a
device, you need to perform a conversion, whether you're displaying Lab on
an RGB monitor, printing Lab to a CMYK press, displaying scanner RGB on an
RGB monitor, displaying CMYK on an RGB monitor, printing scanner RGB to a
CMYK press, etc.


Generally speaking, you'll need to do at least one conversion, from input
space to output space. If you use Lab, you need to do at least two
conversions, one from input space to Lab, one from Lab to output space. In
practice, we often end up doing two conversions anyway, because device
spaces have their own shortcomings as editing spaces since they're generally
non-linear.


The only real advantage Lab offers over tagged RGB is that you don't need to
send a profile with the image. (You do, however, need to know whether it's
D50 or D65 or some other illuminant, and you need to realise that Lab (LH)
isn't the same thing as Lab.) In some workflows, that may be a key
advantage. In many, though, it's a wash.


One thing is certain. When you work in tagged high-bit RGB, you know that
you're working with all the data your capture device could produce. When you
work in Lab, you know that you've already discarded some of that data.

Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Tim Lookingbill
Sr. Member
****
Offline Offline

Posts: 1154



WWW
« Reply #76 on: May 01, 2013, 09:59:26 AM »
ReplyReply

First of all, don't sweat the small stuff. Conversions between different color modes won't destroy your images if you know what you're doing. As there are more than fifty ways to leave your lover, there are more than fifty ways to fiddle the curves in LAB mode. Want to learn more. Read this:

The Jacob’s Ladder Color Correction Technique



also check out this Yahoo discussion thread

Jacob and I don't share the same sense of color balance and color aesthetics just as it's always been every time I see mages edited in Lab space. Folks who work this way are more enamored with their own cleverness in their command of the process over getting color that doesn't look butt ugly or distorted in some way as defined not only from a color constancy aspect but from a color design sensibility that was established 100's of years ago by renaissance artists like Rembrandt and Vermeer onto Kodak color scientist's development of the photo chemical silver halide process.

Now I know why mathematicians, scientists, engineers and your general gear head don't make very good visual artists.
Logged
Gulag
Full Member
***
Offline Offline

Posts: 215


« Reply #77 on: May 01, 2013, 10:33:55 AM »
ReplyReply

It sure could if you're silly enough to do this in 8-bit per color!

A keeper on the sillyness of conversions not needed by the late great Bruce Fraser in response to the Lab conversion craze started years ago. What Bruce wrote then is as true  today. Plus I don't see anything in your link that couldn’t be achieved faster, with no data loss in a good raw converter.


I have seen some great work by those only use 8-bit GIMP as their chosen pixel manipulation tool. What's the orthodox here? Perhaps you can lecture them on 8-bit vs 16-bit as it has turned out pretty religious these days.

For me, JEPG capture works fine since I am very happy with what I can get in camera, and what I can do in post. 16-bit JPEG? I am not so sure?

Jacob and I don't share the same sense of color balance and color aesthetics just as it's always been every time I see mages edited in Lab space. Folks who work this way are more enamored with their own cleverness in their command of the process over getting color that doesn't look butt ugly or distorted in some way as defined not only from a color constancy aspect but from a color design sensibility that was established 100's of years ago by renaissance artists like Rembrandt and Vermeer onto Kodak color scientist's development of the photo chemical silver halide process.

Now I know why mathematicians, scientists, engineers and your general gear head don't make very good visual artists.

You sound like you know what you're doing since you are not one of those who do things by numbers as you mentioned above. 

For me, when it comes down to color inside Photoshop, I can't really see things that others can easily spot. The Info Panel is my guiding star because it reveals helpful numbers in all sorts of revealing ways to help me getting things done, even tho I am no scientist/engineer/math/gear head. I admire those, such as Amy Dresser, who were born with innate ability to tell even slightest change in hue, saturation, and brightness levels just using their naked eyes. You sound like you're one of those who I admire. Does that mean you're better and I am inferior because I have to reply on the numbers revealed by the Info Panel? Secondly, Photoshop gives us more than just three channels natively. Why throw away all the possibilities we're given?
« Last Edit: May 01, 2013, 10:47:25 AM by Gulag » Logged

“For art to be art it has to cure.”  - Alejandro Jodorowsky
Tim Lookingbill
Sr. Member
****
Offline Offline

Posts: 1154



WWW
« Reply #78 on: May 01, 2013, 10:35:26 AM »
ReplyReply

Here's some color corrections I performed on "downsized for web" jpegs in Adobe Camera Raw. The first link below shows how I corrected for the contrast/saturation "distortion" many complain about working in RGB space and end up turning to Lab to fix. I'm well aware of this kind of distortion and thus figured out a simple way that doesn't involve a long drawn out, convoluted editing routine to remedy.

http://photo.net/digital-darkroom-forum/00Toub

This correction of the green rock wall image below was also done in ACR, but even though the poster preferred my version over the others he didn't seem to want the ACR edits which are MUCH easier to convey and post screengrabs for than a bunch of complicated layer blending mode and Lab curve instructions.

http://photo.net/beginner-photography-questions-forum/00YmG4

I'll take those results over what I'ld get doing it in Lab. The color design of the original image was preserved to boot.
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9100



WWW
« Reply #79 on: May 01, 2013, 11:50:11 AM »
ReplyReply

I have seen some great work by those only use 8-bit GIMP as their chosen pixel manipulation tool. What's the orthodox here?

For one, to separate aesthetics of images from the technical realities of image processing in terms of workflow, best image quality, flexibility and so forth. Now do you want to bring up great work which is an aesthetic discussion? Or continue on with the realities of image processing on data and it's effect?
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Pages: « 1 2 3 [4] 5 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad