# Can we predict the color that will be perceived by the eye?

## Main Question or Discussion Point

Hi,my question is the following:

Supposing that we light up the same surface of an optical system (the eye) by two different monochromatic lights (each one characterized by a wavelength and intensity): Can we predict the color that will be perceived by the eye?

Related Other Physics Topics News on Phys.org

Hi,my question is the following:

Supposing that we light up the same surface of an optical system (the eye) by two different monochromatic lights (each one characterized by a wavelength and intensity): Can we predict the color that will be perceived by the eye?

Are you referring to qualia? Whether we can predict the subjective experience of a wavelength?

If so then the answer is no we can't. Although if you have already experienced some monochromatic light source and you then illuminate your eye with another source of similar wavelength then we can say that we will perceive a similar colour, say a different shade of red for example.

Yet, we know that if we light up the eye with two monochromatic lights (let's say blue and red) both of the same intensity, then the color perceived will be purple; if the intensity of blue is more important than the one of red, then the color will be perceived as "dark purple"...
Is it possible to predict the color we see (by telling "x% of blue(435nm) + y% of red (700nm) is equivalent to a light of z nm")?

CONE CELL makes colorful image in blazing light!

Andy Resnick

Hi,my question is the following:

Supposing that we light up the same surface of an optical system (the eye) by two different monochromatic lights (each one characterized by a wavelength and intensity): Can we predict the color that will be perceived by the eye?

Sometimes, but not always. For example, how does one generate pastel hues from monochromatic sources? Or brown?

The answer to this question is also dependant on what else is withing your field of view.

Here's an experiment we did in physics class:

Have a room that can be completely darkened.
Have 3 light sources of variable intensity, one red, one green, one blue.
Have a 6' by 6' grid of 36 different colored squares on a wall.
Have a photometer to measure the light reflected off individual squares.

Now view the grid with an equal intensity of light from the 3 light sources shinning on it. You should see it in it's natural color. Measure that reflected light with the meter and note the measurement, for one of the squares.

If you now cover all but that one square, and adjust the 3 source lights until the color of the 1 square appears different, then remove the covering so that your eyes can compare it to it's neighbouring square it will return in appearance to it's natural color.

You can now measure the light reflecting off that 1 square and confirm that it's peak wavelength differs from when it was under even lighting from the 3 sources.

So, you can in fact see the same color in the 2 cases, when seeing different wavelengths of light.

Q_Goest
Homework Helper
Gold Member

Don't LCD computer screens use 3 monochromatic lights in each pixel?

Andy Resnick

Yes, but the color space accessible by LCD/triphosphor displays is a much smaller subset of the total color space accessible to our eyes.

I find this so interesting. Shine red light and green light into the eye simultaneously, the red light stimulates the red cones, the green light stimuates the green cones, and then the brain's interpretation is "I see yellow." Shine only yellow light into the eyes, it stimulates both the red cones and the green cones, and the the brain's interpretation is "I see yellow." The same sensation in both cases. To find out whether the light is yellow, or whether it's a mixture of red and green, you can't tell by looking at it, although you could tell by directing the light through a prism to see what colors come out the other side. The prism would separate red and green. If it's yellow, it comes out without any change.

This link can also be interesting:http://hyperphysics.phy-astr.gsu.edu/hbase/vision/cie.html#c2";
However; I can't understand what do the numbers in the table represent (eg.Red=191(red),27(green),75(blue)).If anyone has an idea...
In computer graphics, colors are specified with 3 numbers ranging from 0 to 255 with using RGB mode. Those numbers represents the intensity of the colors red, green and blue with three 0s meaning no intensity, hence black. Three 255s would be max intensity for each hence white.

Last edited by a moderator:

And what's the link between these numbers and the chromacity diagram?

And what's the link between these numbers and the chromacity diagram?
There's 2 ways you can specify a color, with a computer display.

1) RGB = red, green, blue

2) HSB = hue, saturation, brightness (alternately HSL or HSV)

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

The wiki entry explains how to convert between the 2.

mgb_phys
Homework Helper

And what's the link between these numbers and the chromacity diagram?
It's called a color space, it defines the coordinates of the R,G,B points on the color triangle and so the range of colors that you can show
- as well as the mapping of RGB->perceived color.
There are two main ones, aRGB (adobe) and sRGB (everyone else) see http://en.wikipedia.org/wiki/SRGB

Thanks a lot for your help that has been very useful to me!
I've one more question:
Is it possible to define the saturation and the hue of a source of light such as a torch or a monochromatic source?

mgb_phys
Homework Helper

A monochromatic source would have a single hue and a sat/value depending on the power.
A black body source (light bulb) is usually defined by color temperature http://en.wikipedia.org/wiki/Color_temperature

It's interesting that lots of stuff that was only of interest to a very small set of optics engineers is now displayed on the back of everyones digital camera.

I've one last question:
Suppose we have two monochromatic light's sources respectively of intensity I1 and I2. Let's say we make the lights of these sources partially reflected by a surface of glass and that x% of the light 1 and y% of the light 2 are reflected:can we say that the intensity of the reflected light is I=x*I1+y*I2?

Q_Goest