Skip navigation

Need lab color info in Camera Raw.

Apr 3, 2011 4:38 AM

  Latest reply: Jeff Schewe, Apr 8, 2011 8:22 PM
Replies 1 2 Previous Next
  • Currently Being Moderated
    Apr 5, 2011 6:55 PM   in reply to nortj

    nortj wrote:

     

    To do fast accurate color correction by the numbers, you need the Lab values. Doing color correction visually is overrated because you can’t calibrate your eyes!

     

    Well, you can calibrate and profile your display so since the odds of ACR giving readouts in a color space it can't convert to is about nil (without a much better use case argument) it's all ya got. You may as well learn how to use the tools you've got rather than hoping for something that is very unlikely.

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 5, 2011 9:46 PM   in reply to Tim Lookingbill

    Looking at it from another perspective (the framers) one can take an image and either destroy it or enhance it by the matte(s) frames and lighting.

     

    We have to develop the notion that a photograph is an object to be looked at, to be treasured for it's own sake, as sculpture, painting or any other object, like an Oriole or Cardinal for instance. Revel in the joyfulness of that, no matter the subject. Make it work.

     

    That applies to many art forms. I once sat down at a Steinway that had just been freshly tuned. Checking out the tuning, I started with some scales then switched to arpeggios and suddenly, I had an appreciative audience.  They just sat and listened, let the sheer joy of the wonderful sound of a Steinway take hold.

     

    How many of us can do that with a photograph? Edward Weston did it; check out Pepper #30. Is it a pepper, or....just a photograph? Irving Penn did it. Look at his platinum prints of cigarette butts, picked up from the street.

     

    Experiencing a photograph is first a revelation,; I am compelled to put every thing else aside and suspend disbelief. If that doesn't happen, I walk away. When it does, the entire content of that image together with the paper itself, the whole becomes greater than the sum of it's parts.

     

    That is the final truth.

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 5, 2011 9:54 PM   in reply to Hudechrome

    A Post script:

     

    My SO read what I posted, and she said....... "except you don't strike me as the type that would like photographs of cigarette butts".

     

    I looked at her and said:

     

    "I rest my case".

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 7, 2011 6:26 AM   in reply to Tai Lao

    Tai Lao wrote:

    ttttt

    As I suspected, it would seem that followers of Dan Margulis are behind this. 

    I have Dan's LAB book and find it highly intestering. I find that the RGB model used by ACR works fine for the vast majority of my work and rarely convert to LAB in Photoshop. However, opponent color models did not originate with Dan, but can be traced back to the German philosopher Goethe in 1810 and expanded by Herring in 1892.

     

    http://en.wikipedia.org/wiki/Opponent_process

     

    Our color photoreceptors (the cones) and digital cameras use the RGB model, but opponency is introduced by the bipolar cells in the retina, with further processing in the optic radiations and visual cortex. The opponent colors are approximately red vs green, blue vs yellow, and black vs white. Red and blue are perceived as magenta, and blue and green produce cyan. RedGreen and YellowBlue are not normally perceived as such, but can be generated under certain conditions (see the Wikipedia article and look up a recent article in Scientific American).

     

    Keeping these facts in mind, it is reasonable to presume that some color edting situations would be more easily done with the opponent LAB model rather than RGB, taking advantage of the psychophysiology of human color perception. LAB readouts and color controls in ACR might be useful in some situations.

     

    Regards,

    Bill

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 7, 2011 1:32 PM   in reply to nortj

    I've scanned negatives and converted to Lab and edited in Photoshop as well.

     

    I was editing some blue pansies surrounded by orangish red (copper colored) bark mulch with light lime green leaves next to neutral looking limestone. The interplay of hues was giving me trouble with my eyesight. I could really get some strange intense hues working in Lab over any other color space.

     

    The problem that I discovered is these hues violated the natural sprectral reflectance relationship of colors within the overall image. Because my eyes became accustom to the look of these gorgeous hues through adaptation inherent within the scene which was shot with the sun low in the sky, I didn't realize how far I went with the color after walking away and coming back to note how weird it looked.

     

    I notice this in quite a few images edited in Lab and posted online in thread discussions from folks who follow Dan's instructions. From my own experience I found working in Lab was not a natural way to manipulate color.

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 7, 2011 1:58 PM   in reply to Tim Lookingbill

    I would hate to tell you the number of deletes I've done the "morning after" over the years! Adams had an interesting phrase covering this, although he meant it with respect to the let down looking at images after the shoot:

     

    Enthusiasm for a subject many times masks a clear concept of it.

     

    Probably not the accurate quote, but close enough.

     

    Morning afters can be a real bummer!

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 7, 2011 5:14 PM   in reply to Bill_Janes

    Bill_Janes wrote:

     

    Tai Lao wrote:

    As I suspected, it would seem that followers of Dan Margulis are behind this. 

    I have Dan's LAB book and find it highly intestering. … However, opponent color models did not originate with Dan, but can be traced back to the German philosopher Goethe in 1810 and expanded by Herring [sic] in 1892…

     

    It's a safe bet to suspect that most photoshoppers correcting colors by the numbers in L*a*b* are more likely to be familiar with (and influenced by) Margulis than Goethe and/or Hering.   

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 7, 2011 11:54 PM   in reply to Tai Lao

    Goethe on color should be better understood because, as opposed to Newton, who made his observations based on the color spectrum, Goethe was concerned with the perception of color. There is a distinction, and individuals concerned with both physics like Werner Heisenberg and mathematics, like Kurt Godel, took an interest in it. Newton studied the diffraction effects by setting a prism on a piece of paper, Goethe did his by holding the prism to his eye.

     

    Of course, Newton won out as his theory could be quantified where Goethe is less so. I don't think Goethe's work could be called a theory in the way Newton's could, so the study of color proceeded from the study of the color spectrum ala Newton.

     

    So here we are trying to deal with the perception because, after all, that's what we do as photographers and painters, but using Newton analysis.

     

    I suspect that the eye/brain perception of color phenomenon is strictly survival based, the need to be able to positively ID objects according to their color was vital, and under varying lighting conditions. Now, we use it as a means of expression beyond basic survival needs, and we find ourselves having to hold certain conditions steady in order to produce work that is consistent from day to day, at least commercially. I for one would, if I could, eschew this as a fundamental condition that must be met, but even producing as an art form, I cannot avoid this influence. So I look to a variety of methods, from RGB to HSL to LAB, even to exotic systems like scene referred.

     

    Sometimes it is a great relief to instead, push the grayscale button!

     

    Message was edited by: Hudechrome

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 8, 2011 7:24 AM   in reply to Hudechrome

    Hudechrome wrote:

     

    …I don't think Goethe's work could be called a theory…

     

    …and Goethe himself did not label it as such.  The term "theory" was introduced in some translations of the original German (Zur Farbenlehre, on the study of colors).

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 8, 2011 9:10 AM   in reply to Hudechrome

    Overall color consistency image to image is the million dollar goal here. Do we define it by the numbers or by our perception? Inconsistent lighting and exposure parameters when shooting a wedding outdoors can skew both especially when defining consistency by something as subjective as "pleasing" color which is what sells.

     

    I've seen wedding photographer's sites where all the outdoor images seem to have a general overall color consistency where it's clear the camera was all over the place shooting backlit into the sun low or just above tree tops, shooting in the shade and mixed lighting temps and type. And some of these galleries were shot as jpegs with very little editing as claimed by the photographer.

     

    I doubt they had the time to check the numbers for consistency and accuracy. IMO I think it's far more difficult to create a consistent "pleasing" color style treatment applied to a series of images shot under a wide range of lighting and exposure setups than it is to go for "accuracy" according to the numbers.

     

    It's especially difficult to achieve both accurate by the numbers and pleasing results on a consistent basis as well.

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 8, 2011 9:56 AM   in reply to Tai Lao

    Let me correct myself here.

     

    I don't think Goethe's work can be called a Scientific Theory. A more general use of the word Theory could encompass Goethe's work, in the sense that we have theological theory or theoria. Synonyms include speculation and conjecture. In mathematics a conjecture is not the same as a theory. Goldbachs  Conjecture is probably the most famous unsolved problems in numbers theory. So, imho, Goethe's work falls somewhere between a theory and a conjecture, leaning pretty close to conjecture. We can all "prove" that he is right by repeating his experiment, but mere observation is insufficient to classify our observation as a theory. OTOH, there are many mathematical conjectures that have been proven or disproved, moving them into either theory or oblivion.

     

    Our problem comes down to this. We can speak of "Red" but to me it's not Red  but reddishness, or greenishness . At some point we can all agree that we are not looking at blue but purple etc.

     

    Anyway, we have come a long way to handling managing color but we are not there yet. I'm not sure there is a there there.

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 8, 2011 9:59 AM   in reply to Tim Lookingbill

    And yet once "pleasing" results are obtained, the numbers are there! RGB, HSL and L*A*B.

     
    |
    Mark as:
  • Currently Being Moderated
    Apr 8, 2011 8:22 PM   in reply to nortj

    nortj wrote:

     

    If you are talking about images from thirty year old negatives where the dye layer curves were not parallel to begin with and the dye layers have faded at different rates, you have a color correction problem that is quite different.

     

     

    And to be precise, Camera Raw was designed first and foremost as a raw capture processing tool. Yes, it can open JPEG and TIFF images but that is not ACR's primary directive. Unless you can make a strong use case for putting Lab readouts into what is designed primarily a raw processing engine, you simply won't get much traction. There's a perfectly good tool for that, it's called Photoshop (which ACR conveniently opens the images for).

     
    |
    Mark as:
1 2 Previous Next
Actions

More Like This

  • Retrieving data ...

Bookmarked By (0)

Answers + Points = Status

  • 10 points awarded for Correct Answers
  • 5 points awarded for Helpful Answers
  • 10,000+ points
  • 1,001-10,000 points
  • 501-1,000 points
  • 5-500 points