Colour

White light is considered to be made up of the primary components Red, Green and Blue. Inks absorb light, and the inks which absorb each of the primary components of light are Cyan, Magenta and Yellow.


fig. 1. The additive colours of light


fig. 2. The subtractive colours of pigments

Since in practice it is difficult to achieve a deep black by mixing pigments of the three primary colours, black ink is typically used together with Cyan, Magenta and Yellow as the four process colours of ink in lithographic printing. These are designated CMYK. The K stands for "Key" and avoids confusion with B for blue.

Throughout the 1980's and 1990's, almost all computers used Cathode Ray Tube displays - the same technology used in televisions. Colours are produced in these types of display by mixing light. Consequently it is usual for colour to be represented in a computer in terms of its Red, Green and Blue components. Like the monochrome images previously considered, a colour image can be represented as a series of pixels. Data is then required to determine the colour of each pixel.

Because the eight-bit byte is convenient, because limits must be applied to the amount of memory used to store an image, and because the human eye does not detect infinitesimal gradations of colour, computer displays typically use an eight-bit byte to control the brightness of each colour component of each pixel. Consequently the colour components of each pixel are determined in a range of 0 to 255.

An image of this type thus requires 24 bits per pixel and can contain around 16.7 million different colours. An image of 800 x 600 pixels will require around 1.4 Megabytes of memory.

The synthesis of colour as a mixture of primary components each of brightness on a scale of 0-255 can be illustrated using a computer graphics package such as Paintshop or MS-Paint:

Exercises:
1) Use Paintshop to create your favourite colour. What are its R, G and B components?
2) How many bytes would be needed for an 8 x 11 inch 24-bit colour image at 1200 dots-per-inch?

CMYK versus RGB

We have seen that the CRT display uses the additive primary colours of Red Green and Blue, and is composed of pixels each of which is the same size. By comparison, printing necessarily uses subtractive colour mixing. Photo-lithography typically uses the process colours of Cyan, Magenta, Yellow and Black inks (CMYK) in full-colour printing, and the density of colour is determined by the size of ink dot produced by screening together with the resultant exposed background white of the paper. This is illustrated by the enlarged photograph of a printed product shown below:

Source: "The Reproduction of Colour", Hunt (5th ed) Fountain Press 1995

An image produced by lithography will thus necessarily differ considerably from its digital equivalent RGB image. The techniques and skills of digital image processing are used to make these differences as far as possible imperceptible to the human eye. For example, an image may be scanned into the computer with the colour of each pixel represented by 12-bit CMYK values - 48 bits per pixel - and handled as CMYK throughout the system. An RGB version of the image would only be produced temporarily to allow the operator to view the image.


Screening

The term screening refers to the technique of using an optical grid (or screen) during photographic exposure. The technique converts an image of varying tone into a series of dots, where areas of lighter tone have smaller dots. When such an image is transferred to a printing plate, the image of varying tones can be reproduced using a single colour of ink.


figure 1 Original Image

figure 2 Detail of Halftone Screen

Source: "Production for the Graphice Designer", Craig, 1978


Colour Separations

Full colour images are typically printed by the offset litho process using four colours of ink. Cyan Magenta and Yellow inks are used to create the range of colours: close juxtaposition of small dots of two or more of these colours of ink will give the eye the impression of the colour formed by mixing these pigments. Some mingling of inks may also take place on the printed page. Note in the example how cyan and yellow inks are combined to produce the blue sky and green grass. In the printing press, paper is presented to successive rollers each of which takes its image from a plate bearing ink of a particular colour.

The colours Cyan Magenta and Yellow, mixed in different proportions, will in theory yield a full spectrum of colours and range of shades. However in practice it is difficult to obtain good dark shades of grey by mixing typical coloured inks; a Black ink is therefore used on a fourth printing plate. Black is represented by K in the initials CMYK.

To determine the intensity of each colour, equating to its proportion in the mixture of colours, screening is used to create dots of different size or density. The extent to which the white paper background is exposed or obscured by the use of smaller or larger dots will determine the lightness or darkness of an area.

The following pictures illustrate the breakdown of an image into four colour separations each of which has a screen applied. The colour separations are re-assembled in the final picture to yield an approximation to the original image. The appearance of this final picture is similar to the appearance of a typical colour printed product (such as a picture in a magazine) when viewed under a microscope.

Click on the links below to see each larger scale image:

Original image Cyan Magenta
Yellow Black Printed Product

An alternative approach is to choose colours of ink which suit the image, rather than simply use process colours. Different shades can be produced by the use of screens.

In this illustration, the original image is composed of 7 colours (in addition to white): yellow (toadstools), red (spots on toadstools, chipmunks' tongues), green (grass), blue (sky), grey (chipmunks coats), brown (log), black (noses, eyes, outlines).
We might decide to print this image in four colours of ink: black, red, sky-blue and yellow. Solid areas will be printed where the colours correspond exactly. For the other areas, screens will be applied. Black with a screen will create the grey of the chipmunks' coats. Blue and yellow will each be screened for the grass, and their juxtaposition in the final product will create a green.

Original Image
Black component: with a screen this creates the grey, and black with a lighter screen contributes to the brown Blue component: with a screen this mixes with yellow to make green
Yellow component: with a screen mixes with blue to make green, and with a lighter screen contributes to the brown Red component: with a light screen this contributes to the brown

Duotone images use two colours of ink, for example black and orange inks can be used to reproduce a photograph in "sepia".

Whilst the use of process colours is common, we may choose to print in any colour of ink, and on any colour of paper (or other substrate). Consequently the separation of the original image into constituent colours need not always be CMYK.