Color is a visual and perceptual neural response to visible light. Visible light is part of the radiant energy (electromagnetic) spectrum, and all radiant energy is essentially the same. The physical difference between radio waves, microwaves, infrared, visible light, ultraviolet, X-rays, and gamma radiation is frequency and wavelength.
Refraction and the Color Spectrum
Light divides into individual colors when it passes through a prism. This is called refraction. When a prism refracts white light, each frequency deflects at a different angle and the beam emerges from the prism as the spectral colors: red, orange, yellow, green, blue. This refraction occurs because the shorter wavelength colors have a shorter wavelength and bend more than the longer wavelength colors.
Wavelength orders the spectral colors, and nanometers, each one billionth of a meter, measure light wavelength. Only the wavelengths from about four hundred nanometers (blue-violet) to about eight hundred nanometers (red) stimulate the human retina to give color vision.
Visible Light Spetrum
Science and industry require precise color definition and classification. Words are imprecise to distinguish and describe color. For example, some of the few words that exclusively name colors are: white, black, gray, yellow, red, violet, blue, green, brown, and others. Most colors require compound names, quantifiers, and suffixes to differentiate colors. Examples are: yellow-green, light green, greenish, and so forth.
To index and organize colors in relation to one another industry uses a triaxial system called color space or color model. We use the term ‘space’ because color data occur in three dimensions. Color vision uses the term tristimulus system
Color organization begins with two small groups: the pure chromatic colors, or hues, and the achromatic colors (white, grays, black). All other colors exist within these two extremes. The number of pure chromatic and achromatic colors is small compared to all color possibilities. The eye differentiates about three hundred colors and about one hundred fifty shades of gray. These two color groups are the limits for the differentiation and organization of all other colors.
Industry uses several color organization systems or color spaces. Below are various color spaces in common use today.
It is impossible to describe the appearance of a color definitively. However, it is possible to describe a color’s appearance in relationship to its environment. The eye differentiates colors according to three criteria: hue, saturation, and brightness (HSB). These criteria together define the relationship between colors. A systematic organization of all colors is therefore possible only in a triaxial system.
In the early 1900s, Albert H. Munsell developed a color model based on human perception. His system terms color’s three attributes as: hue, value, and chroma (HVC). The popularity of Munsell’s color system continues today in industry because it separates the color-independent component, brightness (Munsell value), from hue, and saturation (Munsell chroma). This enables two-dimensional color representation. In addition, it is perceptually uniform. Color distances correspond to perceived differences between the colors. With this system, every color has a specific location that provides an objective system for color notation and communication.
CIE Color Space
In 1931, the International Commission on Illumination (Commission Internationale de l’Eclairage) or CIE established a worldwide color measurement standard, the CIE Chromaticity Chart. The CIE System is a theoretical mathematical model of human color perception. In 1976, the CIE revised the chart to create a more even distribution of colors. The revised chart is the current standard for measuring the color of light.
CIE Chromaticity Diagram
The CIE chromaticity diagram is a two-dimensional graph of hue and saturation. This is a logical point for beginning digital color mixture because additional color values occur by gradually reducing light source intensity (Z) until absolute blackness. The chromaticity chart then becomes a three-dimensional color model which is the CIE Uniform Color Space (CIE-XYZ).
CIE-XYZ is the parent system for nearly all color standards. Variations such as CIE-Lab, CIE-Luv, and so forth, serve different scientific and technical purposes.
The CIE-Lab color space separates color and luminance into discrete color space dimensions because our vision is much more sensitive to light intensity than it is to color. The first component, L, represents lightness. The other two components, a and b, represent a two-dimensional color subspace with +a:red, -a:green, -b:blue, +b:yellow. Because the human visual system is more sensitive to the L component, digital systems allocate more data space to it than to the color components, allowing even more efficient coding of the color values (luminance).
CIE-Luv color space employs the theory that perception codes colors into complementary signals: light-dark, red-green, and yellow-blue. The horizontal u-axis plots red and green values. The vertical v-axis plots yellow and blue values. The L-axis plots the light and dark values, located perpendicularly to the uv-plane.
The Yxy color space derives directly from XYZ and graphs colors in two dimensions independent of luminance. The value Y is identical to the tristimulus value Y in XYZ. CIE names the xy values the chromaticity coordinates and computes them directly from the tristimulus values XYZ.
As with Munsell, the CIE-Yxy separates the achromatic component (Y) from the two chromatic components (xy). Two colors, the same except for luminance, have the same chromatic definition, and therefore the same chromaticity coordinates.
Plotting the Yxy values of colors yields a useful graph known as the CIE chromaticity diagram.
When we convert and plot the xy chromaticity coordinates of the pure wavelengths of the visible spectrum, the resulting points all fall on a horseshoe-shaped line known as the spectrum locus. By definition, since all visible colors comprise mixtures of these pure wavelengths, all visible colors must occur within the boundary formed by this curve.
Connecting the spectrum locus curve endpoints forms a line called the purple line or the purple boundary. Colors on this line are mixtures of pure 380nm (violet) and 770nm (red) light.
An additional color system that is very important to the digital workflow in Kodak’s YCC color space. Developed for the Kodak Photo CD, this color space is device-independent suited for data compression and decompression. YCC is a variant of RGB color space.
Vision is highly sensitive and accurate, however, it is also subjective. Vision and perception vary among individuals, and even change with individual emotional and physical states. Therefore, accurate, consistent color evaluation must account for vision deficiencies resulting from the interaction of the optical system and the mind.
Color vision is part of being human. It is almost certain that no mammals up to primates possess color vision, although it occurs highly developed in birds, fish, reptiles, and insects.
Color blindness occurs in some people, while others see only a limited number of colors. It is estimated that approximately 10% of all males experience some degree of color blindness while females are substantially less susceptible with only about 1% experiencing the condition.
Random retinal impulses and involuntary rapid eye movements (saccadic eye movements) keep the vision system perpetually active. Both are essential to vision. Vision soon fades when an image becomes optically fixed on the retina. Part of the eye movement’s function is to sweep the light pattern over the retina to continually signal the mind the presence of the image. However, overuse fatigues the system, and vision fatigue impairs color judgment. For example, viewing a saturated color for a time causes a second color to appear different as vision fatigue subtracts some of the first color from the image. The reduced sensitivity is called negative after image, and is due to retinal photo pigment bleaching.
Even with the complete absence of light there are random retinal impulses reaching the mind. This continuous background of random activity sets a continuous problem for the mind. The mind must ‘decide’ whether this neural activity is ‘noise’ or information. This internal visual noise increases with age and is partly responsible for the gradual loss of visual discrimination with aging. Visual accuracy and adaptation also decline with aging.
Along with vision deficiencies and variation, external conditions also affect color judgment. Some external variables are: diverse lighting types; different substrates; disparate viewing angles; multiple observers; unconventional illumination angle; and the size and shape of color area.
Because color vision requires sensory data, it is impossible to remember color. We can only compare color. However, visual color comparison is almost impossible unless under exactly the same viewing conditions. For this reason, the graphic arts industry has established the standard for color comparison as a neutral environment with 5000-degree Kelvin illumination. This light temperature appears clear and color-balanced, and is ideal for color comparison.
Quantification and Measurement
Precise standardized measurement is also part of color evaluation. The major color measurement instruments are: Densitometers that compute density, the amount of light absorption of a surface or material. Densitometers precisely measure standard colors used in graphic arts and photography. Colorimeters that measure and compute XYZ color values in a way that models vision, and usually report results in a CIE color space. Colorimeters record all visible colors, but generally not as precisely as densitometers. Spectrophotometers that measure and covert spectral data to a CIE color space. Spectrodensitometers that serve all the functions of a densitometer, colorimeter, and spectrophotometer in one device.
To understand how colors are created with printing requires the understanding that all reproducible colors must first exist in the unprinted white paper. If a color is not present in the sheet of paper initially, it cannot be produced with process inks.
Light filtering occurs with printed ink primaries (yellow, magenta, cyan, plus black). This filtering allows for different proportions of colored light to activate the receptors in the eye, thus stimulating different neural responses for all spectral colors.
Printed color reproduction requires reduction of an original color image into the printing primaries (CMY) and black (K). Industry terms this process as color separation.
Color separation employs the color filter principle. Each color filter and negative film record its complementary color or printer. When printing combines the primary printers of cyan, magenta, and yellow and the black on white paper, the spectral colors subtracted by the filter reappear on the printed page.
For example, the blue filter and negative film records only blue producing the yellow printer separation. Upon plating and printing the yellow printer, the yellow ink on the white paper acts as a filter to stop blue light reflection. The eye sees white paper as equal portions of red, green, and blue light. However, if an ink film layer filters blue light, red and green light combined form the sensation of yellow by the viewer. The filtering principle is why the printing primaries are subtractive filters that control the reproducible colors that exist in white paper.
Visible light divides into the spectral primaries: red, green, and blue. RGB colors are the additive primary colors. The printing primaries (cyan, magenta, yellow) are the subtractive primary colors.
The additive primaries (RGB) and the subtractive primaries (CMY) arrange in two-dimensions where additive primary color is opposite its complementary subtractive primary color.
The additive primaries combine to form white light. Pure subtractive primaries combine on the white paper to form black or the absence of color.
While printed and pigment color comprise subtractive primaries and reflected light, RGB color comprises additive primaries and incidence light.
The color spaces above are device-independent. However, RGB color space is device-dependent different devices detect or reproduce a given RGB value differently, since the color elements (such as phosphors or dyes) and their response to the individual R, G, and B levels vary from manufacturer to manufacturer, or even in the same device over time. So an RGB value does not define the same color across devices without some kind of color management. Also, RGB devices require calibration.
See: Wikipedia RGB Color Model , W3 Schools HTML Colors , and Wikipedia Color Management.