We can see in the visible spectrum between 400 and 700 nanometers. A nanometer is 1 billionth of a meter.
The human eye has an imperfect lens that adjusts focus automatically depending on the depth of the objects perceived. The lens in the eye does not correct for chromatic aberration: This means that if we focus on a red object, a blue object will be out of focus by more than a diopter. Note: a 1 diopter lens is one that focuses at a distance of one meter. For this reason pure blue should not be used for displaying text - it will be very difficult to read because it is out of focus on the retina.
Light is captured at the back of the eye by photo-receptors.
These turn light energy in to electrical energy. These photo receptors
are concentrated in a small area of the retina called the fovea.
Our ability to see detail and color is much better in this area
than in other parts of the retina. Powerful muscles cause eye
movements that rotate the eye and allow the fovea to image different
parts of the environment. The fovea subtends an angle of approximately
2 deg. of the visual field.
There are two type of receptors at the back of the eye called.
Rods and Cones. Rods are only active at low light levels and there
are more of them at the periphery of the visual field. Cones are
responsible for color vision and they are concentrated at the
fovea. There are three cone types, each containing a different
photopigment and sensitive to long wavlength, medium wavlength
and short wavelength light respectively. A Cone sensitivity function
is the way a particular cone type varies its sensitivity with
wavelength.
About 10% of the male population are color blind and about 1% of the female population suffers from some form of color blindness. Usually this means that they functionally lack either the red cones (long wavelength) or the green cones (medium wavlength).
The Commission Internationale de L'Eclairage (CIE) is responsible for maintaining color standards. This system is based on the concept of a standard observer. This standard observer is in turn based on a model of the human rods and cones. However, the model does not take adaptation or simultaneous contrast into account which is why the CIE system has little to do with the appearance of colors.
Because light and dark information is more important to our
perception than color information it is conventient to have a
measure of the amount of light emitted by something, as opposed
to the color of the light. Luminance is defined by a Luminance
sensitivity function and this is part of the CIE standard observer.
This is an overall function of the human sensitivity to light
as it varies with wavelength.
In order to determine the luminance of a source we integrate luminance
sensitivity function with the wavelength distribution of a light
source. Because humans are much less sensitive to blue light than
to green light, blue wavelengths contribute much less to the luminance
computation.
In the CIE system color is defined in by tristimulus values,
X,Y and Z. Y is the same as luminance. The tristimulus values
are obtained by integrating the spectral distribution of a source
with the three tristimulus functions. To convert from tristimulus
values to the RGB values of a monitor a matrix multiplication
is required. To obtain the matrix, the monitor phosphors must
be calibrated in terms of tristimulus values.
The CIE chromatiticity order to convert color space into a
luminance component and two chromatic components. Chromaticity
coordinates are defined as follows.
x = X/(X+Y+Z)
y = Y/(X+Y+Z)
Y=Y
There is a special diagram used to plot these colors called the
CIE chromaticity diagram.
A Uniform color space is a space in which equal measured distances
correspond to equal percptual distances between colors. CIElab
and CIEluv are two standard Uniform color spaces
used in industry.
Most computer graphics uses simple transfromations of the R,G,B
values of the color monitor in order to specify colors.
RGB cube
HSV : = Hue, Saturation and Value.
RGYB
Almost all useful transformations differentiate between the Luminance
dimension of color and the Chrominance dimensions of color.
There are two ways of specifying the two Chrominance Dimensions.
1) Hue and Saturation: Hue is the spectral value, Saturation is
the vividness.
Highly saturated colors are "pure". Low saturation colors
are close to the grey scale.
2) Color Opponent: R-G, Y-B (see below)
(Hurvich & Jameson)
Opponent process theory holds that the input from the cones gets
processed into three distinct channels immediately
after the receptors.
It is a fact that we are much better at perceiving most kinds of information if it is presented on the luminance channel.
When you wish to present any of the above kinds of information
make sure that there is considerable luminance contrast. It is
much more difficult to perceive these kinds of information if
they are presented using purely chromatic differences. For example,
using red letters on a green background, where the letters are
equiluminant with the background, will result in something that
is very hard to read.
Color coding (chromatic coding) is good for showing type information.
Make all objects of type X have the same color, that way they
are easily recognizible. This is because in the real world color
mostly tells us about the properties of objects. E.g. is this
fruit ripe or rotten.
In normal environments, one of the major tasks of the visual
system is to determine the surface color of objects under a vast
range of lighting conditions. The amount of light in a candle-lit
room and on a bright day at the beach may vary by a factor of
10,000. Somehow, the brain must still determine what are white
and black surfaces under these very different conditions. Similarly,
the color of the light in the envonment may change a lot but the
brain must somehow interpret the color of the surface despite
the change in illumination.
This ability to perceive surface lightness and color under a wide
range of illumination conditions is called lightness and color
constancy respectively.
There are two main low level mechanisms responsible for constancy:
Adaptation and contrast.
Adaptation (change in gain)
The cone receptors in the eye become less sensitive because of
chemical bleaching when there is a lot of light around. This results
in a reduction in sensitivity. If the light is strongly colored
then the different cone types will become differentially adapted.
In red light long wavelength cones will become less sensitive.
The effect of adaptation is to make the eye have a sensitivity
range appropriate to the environment.
Simultaneous Contrast (lateral inhibition)
There may be a wide range of illumination levels within a single
environment. Something in shadow will be much less illuminated
than something in bright sunlight. Simultaneous contrast is a
phenomenon whereby a grey with a black surround will appear darker
than a grey with a white surround.
The neural mechanism of simultaneous contrast is called lateral
inhibition. Lateral inhibition means that adjacent neurons inhibit
one another. The effect is that the eye is less sensitive to regions
next a bright patch and more sensititve to regions next to a dark
patch. It is a kind of edge enhancing process.
Both luminance and colour contrast occurs.