Photoshop histogram questions

[Abitconfused]

In Photoshop and Camera Raw which histogram display is actually adjusted for perceptions of brightness?

[Manfred M]

Neither as Photoshop and ACR both show essentially the same data - all the histogram shows you is the distribution of the values in each colour channel and the distribution of luminosity values (I'm not sure how each of the individual colour channels are mapped into luminosity histogram). The left hand edge is where there is a "0" value and the right hand edge is "255".

Perceptions of brightness? I'm not quite sure what you are looking for here.

[Abitconfused]

I was thinking that the Luminosity histogram was adjusted for human perceptions of brightness in color and that the Color histogram was not. However, upon reading your reply and further investigation, I see the error of my ways. They seem identical...generally.



and



I still seem to recall that luminance and luminosity are different things, luminosity having to do with perceptions of brightness rather than luminance that measures light in a strictly linear way. Would there be any utility to a histogram that was adjusted for human perceptions of the brightness of various colors?

[Abitconfused]

But then, I have this: "The human eye’s perception of brightness is biased toward green (59%), then red (30%), and, lastly, blue (11%). And I looked at the histogram with a more colorful image, and got these histograms....



and




Which seems to make sense because, while there is a lot of blue in the image's color, from a luminosity standpoint the histogram dutifully disregards it as a significant source of perceived brightness. Which is redundant because brightness is a perception.

[xpatUSA]

Neither as Photoshop and ACR both show essentially the same data - all the histogram shows you is the distribution of the values in each colour channel and the distribution of luminosity values (I'm not sure how each of the individual colour channels are mapped into luminosity histogram).

A luminosity histogram is one of whole pixel values, not split into separate color channels. Each pixel is converted to luminosity by approximately (0.3R+0.6G+0.1B). Thus, 0-255 in the histogram represents pixel gray-scale values even though the image is normally RGB color. The luminosity histogram has it's uses but can not always show a blown color channel, if we think about it.

Perceptions of brightness? I'm not quite sure what you are looking for here.

I think the reference is to luminosity as described above. [edit]Ed confirmed while I was typing this post. [/edit]

Here's an overexposed shot in RawTherapee:



Note that the blue channel in the sky is blown as shown by the black warning. In the histogram, the blown blue graph is clearly indicated but the luminosity graph shows almost a perfect ETTR.

[Manfred M]

(0.3R+0.6G+0.1B).

Thanks Ted - that was the information I wasn't sure about. There is a small typo there, the second "R" should be "B".

Regardless, going back to Ed's original question, none of them are adjusted for "perceptions of brightness", they just show the actuals. Often, when the luminosity histogram has values from 0 to 255, (in ACR / Photoshop) I find I have to adjust the midpoint to get the overall image brightness to where I want it to be.

[DanK]

Regardless, going back to Ed's original question, none of them are adjusted for "perceptions of brightness", they just show the actuals. Often, when the luminosity histogram has values from 0 to 255, (in ACR / Photoshop) I find I have to adjust the midpoint to get the overall image brightness to where I want it to be.

I was going to write the same thing about the individual-color histograms. They are just a display of the data, not an estimate of how the data will be perceived. However, I don't think this is entirely true of the luminance histogram. The weights on the individual channels that Ted described are a response to average differences in how people perceive the three colors. This is noted in the tutorial on this site:

Luminosity takes into account the fact that the human eye is more sensitive to green light than red or blue light.

Ted, you wrote this about your example:

Note that the blue channel in the sky is blown as shown by the black warning. In the histogram, the blown blue graph is clearly indicated but the luminosity graph shows almost a perfect ETTR.

This often happens. I experience it most often with reds rather than blues, e.g., in flower photos, but the general problem is the same. This is why I always have my camera set to show the histograms of all three color channels, and unless I am content losing detail, I set ETTR based on the right-most of the three.

[xpatUSA]

Thanks Ted - that was the information I wasn't sure about. There is a small typo there, the second "R" should be "B".

Thanks Manfred - fixed that.

Regardless, going back to Ed's original question, none of them are adjusted for "perceptions of brightness", they just show the actuals. Often, when the luminosity histogram has values from 0 to 255, (in ACR / Photoshop) I find I have to adjust the midpoint to get the overall image brightness to where I want it to be.

Indeed.

Ed might find this of interest:

http://www.scantips.com/lights/histograms.html
.

[Manfred M]

Thanks for posting the link to the article, Ted.

The important point that is made is: "The histogram is NOT a light meter. It shows where the image tones came out, but it knows absolutely nothing about how they should have come out. White or bright things will appear to the right, and black or dark things appear to the left, but just where they ought to be is simply not known here. It really depends on the color and reflectivity of the subject. That's the photographers job to control exposure. We learn a lot through experience, but we should judge the actual picture."

[Abitconfused]

xpatUSA, thank you for the article suggestion. There is a lot there. The author says this, in part: "Our camera image is RGB color, but when the camera is showing only the single gray histogram, it is showing luminosity, an artificial number, which is not the actual RGB data - it is the computed grayscale brightness equivalent of the colors, perceived when viewing this image on a B&W monitor screen or print. Green looks brighter to human eyes, and blue looks darker, and grayscale should come out that way (B&W film does). This Luminosity formula computes the grayscale difference, by "weighting" the RGB components differently, to change to brightness as the eye perceives them in grayscale. The idea is about which picture grayscale tone properly shows red lipstick or green grass, etc?" Isn't perceived grayscale brightness still perceived brightness however determined?

[xpatUSA]

Isn't perceived grayscale brightness still perceived brightness however determined?

Keeping it simple, "Yes".

Unfortunately though, "brightness" has many meanings so it's use can give rise to misunderstanding.

In particular, the Wiki says this of brightness:

"In the RGB color space, brightness can be thought of as the arithmetic mean μ of the red, green, and blue color coordinates"

which bears no resemblance to the commonly accepted definition of luminosity or luminance - and the formula I quoted for those is defined basically by the response of the human eye to visible wavelength.

So the pedantic answer to your question is "No".

[Abitconfused]

OK, here is the Cambridge in colour description, from the website: LUMINANCE HISTOGRAMS

Luminance* histograms are more accurate than RGB histograms at describing the perceived brightness distribution or "luminosity" within an image. Luminosity takes into account the fact that the human eye is more sensitive to green light than red or blue light. View the above example again for each color and you will see that the green intensity levels within the image are most representative of the brightness distribution for the full color image. This also reflected by the fact that the luminance histogram also matches the green histogram more than any other color. Luminosity correctly predicts that the following stepped gradient gradually increases in lightness, whereas a simple addition of each RGB value would give the same intensity at each rectangle." Is this correct?

[Manfred M]

Is this correct?

Yes.

The histograms showing the individual channels are important in the way that Dan has already mentioned, it shows where we have data loss in these individual channels at the dark end and the light end. That is really the only useful information I get out of them and I only really worry about that while shooting, less so when processing.

The luminance histogram is something I pay a lot of attention to, especially in PP work as I will set the black point and white point based on what I see there, usually, but not always, paying attention how much is clipped. A typical recommendation is to set these to where the data starts to ramp up, as this gets you a good contrast level in the image and in the type of shooting I do, a tiny bit of clipping in one of the colour channels is something I often don't have to worry about.

Once those are set, I will use the mid point slider to set the level of brightness in my images.

[Abitconfused]

Thanks!