The human eye can distinguish about 380 000 colors. The color image is represented by using color models and color indexing. A color model is usually a specification of a 3D or 2D color coordinate system, where the coordinates represent different colors. A color is always represented by three components. These models can be represented by an infinite number of different colors. An image, no matter what the original format is (GIF,TIFF..), is always converted to a color model format, usually RGB, before it is displayed or printed. There are many color models and below are a just a few of them:
RGB is an additive color model, used with PC monitors, cameras and scanners. It represents a color by adding three intensity components together : Red, Green and Blue. The RGB model is a unit cube with the three components along the X, Y, and Z -axis. An image is usually converted to RGB values in PC-monitors before being displayed onto the screen. The word 'additive' in an additive color model like RGB, means that the colors are emitting light (= light source). The RGB system is the most common one.
RGB color model pictures from two different viewpoints.
HSV is short for "Hue, Saturation and Value". The user specifies a color (=hue) and then adds white (=saturation) or Black (=value). This way all kinds of colors can be represented and the users (specially the artists) can define exactly what kind of colors they want. That's why this system is said to be more user-oriented than RGB. The HSV is used widely in image manipulation.
HSV color model diagram
Cyan, Magenta and Yellow are the three components of CMY. A color is specified by removing these components from white (1,1,1) light instead of adding components to blackness, like in RGB. This system is used in color printers with addition of black color (cyan, yellow and magenta produces black gray, not pure black). This is a substrative color system which means that the colors are reflecting light. This system is used mainly in printers.
RGB to CMY and CMY to RGB - conversions:
C |1| |R| R |1| |C| M = |1|- |G| G = |1|- |M| Y |1| |B| B |1| |Y|
The CIE is developed by the International Commission of Illumation. All the colors are represented by virtual colors A, B and C. The three components of any real color are calculated from the following equations:
x = A/(A+B+C) y = B/(A+B+C) z = C/(A+B+C) x + y + z = 1.
In reality, we can take x and y from diagrams (or tables) (see below) and calculate the z-component (= 1 - x - y).
CIE chromaticity diagram, the X and Y-axis are in nanometers.
The dot in the middle is the virtual color C,which is white.
Here is the whole diagram in GIF - format
YUV contains luminance (Y) and two chrominance (U & V) components. The Y-component contains the brightness information (=black and white - image) of an image and the U and V components represent the color information. The human eye cannot distinguish colors as good as gray variations and therefore the resolution of two chrominance components (=color resolution) doesn't necessarily have to be as good as the luminance resolution, like in RGB. Usually either two adjacent horizontal luminance pixels have the same color value (YUV 4:2:2) or 2*2 block of luminance pixels have the same color value (YUV 4:1:1). This way the image can be compressed from RGB 24 bits/pixel to 16 (4:2:2) or 12 (4:1:1) bits/pixel. Even with YUV 4:1:1, the luminance and the color quality is rather equal to that of RGB.
Two links about color:
