Take the 2-minute tour ×
Graphic Design Stack Exchange is a question and answer site for Graphic Design professionals, students, and enthusiasts. It's 100% free, no registration required.

If I convert hundred percent green and blue to grayscale then blue one becomes like twice as dark.

I can see that green looks brighter, but I also want to understand reason behind it.

Looking at colors in rgb model:

  • pure blue = 0 red 0 green 255 blue
  • pure green = 0 red 255 green 0 blue

So why is pure green light is brighter than pure blue?

share|improve this question
2  
How did you "convert to greyscale"? With just the menu command? If you want them to be more accurately converted, use the Black and White command or an adjustment layer which removed all saturation. –  Scott Aug 6 at 21:48
    
yep that was it.._+1 –  Muhammad Umer Aug 6 at 22:05
5  
Short answer: it's called "relative luminance" aka "perceived luminance". Our eyes are more sensitive to some hues (particularly green) so they seem lighter, and photoshop's grayscale function weights hues according to their luminance. I can't find any explanations that aren't horrifically technical... here's one nbdtech.com/Blog/archive/2008/04/27/… –  user568458 Aug 6 at 22:20
2  
Muhammad, about half of your 46 questions have no accepted answer. Maybe you could work on that. –  Joonas Aug 7 at 12:51
    
dont worry about me...i dont accept answer until i think it's a right answer i don't wanna mislead others who stumble upon it searching for the same thing. –  Muhammad Umer Aug 7 at 17:28

5 Answers 5

up vote 29 down vote accepted

Human perception isn't the same for all colors. Our eyes have different color pigments which absorb different frequencies of light.

There's a bit about this over in Physics.SE: Why do green lasers appear brighter and stronger than red and blue lasers? From this question a chart is presented that shows the absorption of different frequencies of light. The three cones are represented by their colors and the rods are the dotted line.

Note here that that something that is in the blue range, while it will trigger the blue cones, but poorly trigger the rods or the green or red cones. Meanwhile, something that is deep red will trigger the red and a bit of the green cone, but not much of the rod or any of the blue cone. The sum of these pigments absorption peaks in the green making green look the brightest.

There's even more about this in the cone sensitivity curve found at Visible Light and the Eye's Response and the spectral sensitivity wikipedia page. When looking at these graphs take care to note that many of them have been normalized into a 0 .. 100 scale.

There is also an aspect of distribution of the various color cells. More on this can be read about in "Blue" Cone Distinctions at HyperPhysics which notes that outside of the Forvea Centralis (most densely packed area of the retina for maximum sensitivity) they are only about 2% of the total count of cone cells (which then suggests that there is a 'blue amplifier' in our perception of the signal). The reasoning for this is likely because the blue light will get bent differently than red or green and we would have significant issues with chromatic aberration in our own eyes if we were to sample blue outside of the main focus significantly (just thinking about that gives me a headache). The root of the color vision tree at HyperPhysics is Color Vision Concepts which is a good read.

This predisposition for green being the important color can be seen in the various Bayer filters used in digital cameras which sample green more than red or blue.

Bryce Bayer's patent (U.S. Patent No. 3,971,065) in 1976 called the green photosensors luminance-sensitive elements and the red and blue ones chrominance-sensitive elements. He used twice as many green elements as red or blue to mimic the physiology of the human eye. The luminance perception of the human retina uses M and L cone cells combined, during daylight vision, which are most sensitive to green light.

From Wikipedia Bayer Filter: Explanation

This can again be seen being utilized in the program ppmtopgm which converts a color image to grayscale. It uses the formula: l = .299 r + .587 g + .114 b - note that green is more than twice as the other colors combined when creating the gray value, and 5x more than blue. Btw, be sure to read the quote in that man page... its quite amusing - especially when taken in context of its origins)

And thus, why green appears to be brighter than the other colors.

share|improve this answer
    
As a possible anecdotal proof: when spinning my sons toy that has seemingly equal shades of red, yellow and blue around it at high speeds, it's almost impossible to notice the blue, you just see yellow-red-yellow-red - there is a gap but it's not obviously blue, just dark. Great answer, welcome to GD SE! :) –  Mr E. Upvoter Aug 7 at 13:23
    
this is so weird..that i knew this and once made a filter in html5 canvas that turned photo into black and white and i used the exact same formula that some website said is best. (my bookmarks need their own search engine :D) ... –  Muhammad Umer Aug 7 at 17:40

"Greyscale" is a print specification basically. Yes it removes color, but the "Greyscale" mode is only really needed for printing. Everything on screen is RGB even if it looks grey. In this respect, when you use the Mode > Greyscale command, Photoshop ignores the RGB data and relies on CMYK color data and your color profile settings to convert.

If you hover your cursor over the colors and look at the Info Panel you'll see the wildly different CMYK colors:

You can see that one color is 69C and 100Y and the other is 90C and 72M (based on my profile settings). Mixing C and Y is always going to be lighter than mixing C and M.
(also note the ! next to the values indicating that the color is actually out of the CMYk gamut.)

This is a simplified explanation, as others have pointed out luminosity is a large factor as well in the conversion.

If you are strictly working with RGB documents, greyscale is a poor choice. You are better served with a Hue and Saturation Adjustment layer which simply removes all saturation -- this results in the same grey for both colors if that's what you want.

You can also use Image > Adjustments > Black & White to convert to a grey image (still in RGB color space though). This allows you to tweak the levels of the colors specifically to achieve the greys you want.

Essentially you're comparing RGB and Greyscale when they really don't have any correlation.

share|improve this answer
    
so in reality they have correlation but not for us humans right?...as our eyes pick up colors differentlty. –  Muhammad Umer Aug 7 at 17:42
    
Well Michael answered the biology behind human perception, which is all true, but has very little to do with your actual question as to why the colors are different when you convert to greyscale. –  Scott Aug 7 at 17:44
    
which sense to me that in reality to mimic luminance as seen by eye that's actually a correct way to do it. As green seen as brighter by eye than blue. Now i know if i want to preserve the color values then i use the mode i used...and i want to convert them as they are in reality i use what you told me. –  Muhammad Umer Aug 7 at 17:53
    
RGB and "Grayscale" have a very concrete mathematical correlation depending on the definition of "Grayscale" being used. The OP is asking why, in the function grayscale(R,G,B) (in his particular case) with a scalar result, grayscale(0,1,0) differed from grayscale(0,0,1). The question was not about the details of how to present a converted color on an RGB display. –  Jason C Aug 8 at 4:49

I find that's it's easier to evaluate RGB colors in the HSB model (Hue, Saturation, Brightness). In this case, it helps you in two ways.

enter image description here
1. It helps me remember that some colors have greater inherent luminance/brightness than others. Just take a look at the green and blue points on the Hue scale. Green is clearly brighter.

enter image description here
2. If you're looking to control the value levels, you have a clear indication in the brightness scale (once you remove the saturation variable).

share|improve this answer

The best way to inspect the differences in lightness is to use the Lab colour space, rather than RGB or CMYK.

The Lab colour space was developed to be a better representation of human vision than the RGB colour space. The L channel in particular, which stands for "lightness" and is the relevant channel for this discussion, attempts to be a closer match of human perception of lightness.

If you compare a pure green vs pure blue image, you'll see that the lightness channel has different values. While pure blue has a value of 30 (out of 100), green has a value of 88.

Lightness values

share|improve this answer

Its because of the function used internally when converting to grayscale. Its a weighted ratio addition of the three components, with more weight assigned to green compared to blue.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.