Understanding Colour

Part 1: Computer Images

This feature helps you to understand how colours are represented and how images are stored on computer systems. If you use Photoshop or a similar imaging program, or a digital camera, it will help you understand and use the settings in the software and camera menus. You will find some of the ideas here essential to getting your prints right - the subject of a later feature.

Pixels
Images displayed on computer screens are composed of lines of coloured dots we call pixels (short for 'picture elements'.) On a typical computer, each of these pixels is made up of three dots, one red, one green and one blue (RGB). By varying the intensity of these three dots, a wide range of colours can be produced. The obvious way to store an image in a computer system is simply a list of these values for each dot in order, pixel by pixel and line by line. At the start of the file there also needs to be some other data about the file, perhaps for example giving the lengths of the line.

Bitmaps
Files such as this are the basis for most image file formats on computers, and they are generally called 'bitmaps', 'paint' or 'raster image' files. We can safely leave the more intimate details of their formats to those who write software to handle them, but we will sometimes need to know about the types of image data each file format can handle. Common examples of such formats include BMP, PICT, PCX and GIF.

Vector Files
Images can also be represented on computers in a totally different way. Rather than storing the actual image as dots, software can store a set of instructions for drawing an image on a screen (or other device.) These instructions can be to draw lines from one point of the screen to another, to fill in areas with a particular colour, and other more complex actions. To make the picture, these instructions have to be interpreted by suitable software. Images that are stored in this way are called 'vector' files. Examples would include Corel Draw CDR files and Postscript files of various types. Vector files can also include some areas as bitmaps, and are thus in some ways more versatile.
Vector files have other advantages. Generally they are smaller in size than bitmap files. Also the resolution of the image that is produced is not generally limited by the file, but by the screen or printer used to output the file. Theoretically at least it can get the best possible result from the screen or printer, so long as the software used to interpret it is doing its job as well as possible.

Worth 20 Thousand Words?
For our normal 24 bit colour display, each pixel is represented in the file by 3 values, each of 8 bits, corresponding to the RGB dots. Since 8 bits is one byte, there are three bytes for each pixel. This means image files are relatively large; even a small on screen image, perhaps 300x200 pixels, contains 60,000 pixels and would need 180,000 bytes.
A picture may be worth a thousand words, but typically a text file this size might hold 20,000 words! For many purposes simple files such as this are too large, and formats that can compress image data - such as JPEG are used. Other formats are also often compressed.
Some file formats are more versatile than others, and there are many different 'varieties' for example of TIFF files. For the user of image files, the important aspects of any image file (apart from its ability to be written and read by software) are the way it represents colour (RGB in the example above), the number of bits it can handle and the type and amount of compression it allows. You will find more detail on this in Part 6 of this feature (see box, top right.)



Part 2: Colour Spaces

CIE
In 1931, an international committee, the CIE, met in Cambridge, England and defined a method of describing all possible colours that could be made by mixing red, green or blue light sources. They showed these as a two dimensional graph. This showed the hue (colour) and chroma (saturation or strength of colour). A third element or any colour, its lightness or luminance, was represented by a third dimension not shown in the simple diagram.
The 'Commission Internationale de l'Eclairage' (CIE) representation gives a horseshoe like curve representing the pure spectral colours, from violet (4000A) at the bottom left around through blue, green, yellow and orange to red at the other end of the horseshoe. A straight line joining the two ends adds the various mauves, mixtures of blue and red, as shown in the rough diagram of the CIE curve below.



















Somewhere in the middle of the horseshoe is a vital point, the white point. A line from here to any point on the curve represents colours with identical hue but different colour intensity or saturation. So, for example, the line shown goes from white in the middle of the diagrams to yellow at the edge, all the points along it are the same yellow hue, but getting paler towards the white point. The saturation of the colour increases towards its maximum on the spectral curve.
Of course on a monitor we cannot actually show all of the possible yellows and the colours here are only to demonstrate the idea.

CIE-LAB
The CIE definition was revised in 1976 to produce a more uniform spacing of colours which observers were able to distinguish, producing the CIE LAB colour space (properly CIE L*a*b*), capable of representing all possible colours. This uses three variables, a luminance, L* (L-star) and colour values on a red-green axis (a*) and a blue-yellow axis (b*).
Lab colour is hard to understand, and although it provides the widest representation of how we see colours, it is not a good match to either how input devices such as cameras or scanners represent colours, or how output devices - printers, monitors etc - reproduce them. Its importance is as a reference space that is independent of actual devices, and it is widely used as such in colour management and for changing between different colour spaces. It is the basic colour model around which software such as Photoshop was built.
It often comes as a surprise to photographers to find that there are many colours that cannot be produced on film or print. The range of colours that a particular medium can produce is called its gamut. Colours that cannot be shown in a particular medium are said to be out of gamut. How these differences in gamut are handled is specified by 'rendering intents' which will be dealt with in a feature on printing.



Part 3: RGB & CMYK

Alternatives to LAB
Lab colour is important as a reference space, but in working with colour it is generally more useful to work in spaces that are connected to actual colour output, whether on screen or in print.
The monitor display, using illuminated red, green and blue dots has a different range of colours to a photographic print or the printed page (RGB). Photographic colour prints rely on cyan, magenta, and yellow dyes formed in the emulsion of the photo paper during development (or in the case of Ilfochrome, on those that remain after some are destroyed in development.) With non-photographic colour printing, a black ink is also needed to give the images sufficient depth. Alternative ways of representing colour are used which correspond to these different methods of reproducing colour, called RGB and CMYK respectively.
Unlike Lab, which is absolutely defined, these colour spaces are device dependent, and images using them are altered, largely in brightness, by the gamma of the system displaying them.

RGB
RGB colour spaces represent colours in terms of values from 0-255 representing the nominal red, green and blue values required to produce them. CMYK uses values for the amounts of the different inks. There are many different RGB (and CMYK) spaces, each of which has a different gamut, but you should choose one of the standard spaces unless you have good reason to do otherwise.
For web use, the recommended RGB colour space is sRGB, and for printing, we most often use the Adobe RGB (1998). Other possibilities include Apple RGB, now mainly of historical interest, and ColourMatch RGB, which once used to be a standard for commercial CMYK printing. However, Adobe RGB (1998), has the advantage of a larger gamut, and is now becoming a more widely accepted standard for RGB files for printing purposes. Both Adobe RGB and sRGB assume a gamma of 2.2, the normal (calibrated) figure for PC screens (see the feature on monitor calibration, box at top right, for more on gamma.) Apple RGB used gamma 1.8, and will thus display images looking lighter. An article by Dan Margulis, 'Fate and the False profile' shows how you can make good use of incorrect RGB profiles.

CMYK
CMYK is more complex, as it has to take into account the different properties of inks and papers used for printing. In general it seldom makes sense for photographers to work with CMYK files, but to leave the conversion from RGB to CMYK to the print designer or printer. If you are asked to provide CMYK files you will need to find exactly what is required from your printer.
CMYK can sometimes be useful for balancing colours in images that are otherwise difficult to get right. Dan Margulis, in features and books, such as his various editions of 'Professional Photoshop' goes into this approach in some detail. There is a chapter of this on the web which will give you a good idea of his approach. If you are involved with commercial printing and want to know more about how this works - or have to make your own CMYK conversions, his books are worth detailed study.

Changing spaces
Different colour spaces have different gamuts, and when changing from one to another, some colour information will be lost because it cannot be represented in the new space. So normally you should only change from one space to another for very good reason. LAB space has a larger gamut than others, so there no loss in going to that from another colour space, so long as the software uses a sufficiently accurate representation of lab space.
Both RGB and CMYK can only represent at best a relatively small part of the total colour space, and there are some colours that can be produced in one but not in the other and vice-versa. In general, there are often some yellows and possibly blue-greens that can be printed but not shown on a monitor, while monitors can usually display more intense greens, reds, blues and magentas than can be produced by normal four colour printing. Printing also generally allows you to get the impression of a deeper black than a monitor can provide.

Soft Proofs
Photoshop in particular allows you to preview the effect of printing by making use of its 'soft proof' feature. In the View menu, you can use the 'Proof Setup' option to load a profile for the printer colour space. You can then toggle between your RGB working space and this other space using Ctrl and Y (or Command and Y.)
If you are printing to an inkjet printer, you will send data to the printer as an RGB file, and the printer driver will carry out the conversion needed for your inkset - whether the conventional CMYK of 4 colour printers, or more complex six or seven ink solutions. Your printer profile file allows you to soft proof before printing as above.
Colour management and the different rendering intents that can be used in the conversion between RGB and CMYK (and other colour spaces) will be dealt with in detail in another feature on printing images.



Part 4: Other Colour Modes

sRGB
The standard space used for images on the web is sRGB. This has a significantly more limited gamut than Adobe RGB (1998), and should normally be avoided whenever possible for other uses. Unfortunately it is currently the default colour space for Photoshop (although very easily changed - Edit menu, Colour settings), and also for many digital cameras where it is often the only space supported. Professional digital cameras can usually also use Adobe RGB (1998), and should be welded at that setting for most purposes. sRGB does not allow many intense greens and blue greens to be shown. If you are choosing a camera for serious photography, the ability to work in Adobe RGB (1998) mode is a point well worth checking.

Indexed Color
Photoshop also has other modes for working with images that allow a more limited range of colours. Indexed color is mainly of interest in preparing non-photographic graphics for web use. It allows you to select a limited range of colours to be used in a particular image, usually between 3 and 256. When working on a True Colour display, these can be chosen from the full range of 16 million colours, enabling a fairly accurate choice of colour for many graphics. Other colours can then be produced from the 256 by using patterns of different colour dots, a process known as dithering, although this may result in some loss of sharpness.
Indexed colour seldom gives good results with photographs. Its advantage is in reducing file sizes, as 256 colours can be stored in 1 byte, rather than the 3 bytes needed for a normal 24 bit image. For many simple graphics it is possible to reduce the number of colours still further; if 8 colours are enough, then only three bits are needed per pixel, reducing the size compared to the 24 bit image by a factor of eight. Such reductions are important when considering the loading speed of web pages.

Grayscale
Black and white images can be reproduced either as 24 bit colour or as grayscale images. In the 16 million colours of 24 bit colour there are only 256 pure grey shades (254 greys, black and white), and making use only of these allows a grayscale image to be stored as one byte per pixel.
However, black and white images are seldom purely neutral, often having an overall warm or cold black colour, sometimes more complex, with different colour nuances in different tones. Some may even be definitely sepia or otherwise tones. These effects are lost in changing to grayscale, and photographs may appear less rich. If you are preparing images for the web, it is often worth leaving them as full colour rather than changing to grayscale, and always worth checking.
Some ink jets allow you to choose between printing in colour - using all the inks - or simply using black ink when printing a grayscale black and white image. Almost always the print made using the full set of inks will result in a richer and less 'dotty' print. However it is often very difficult to get a print that looks adequately neutral using colour inks. If you are really devoted to black and white printing, a better solution may be to set aside a printer for that purpose, using a set of special black inks in place of the normal colour inks. These solutions will be looked at in more detail in a later feature.

Duotone
The photographs in some books are reproduced to a very high standard (sometimes one difficult to reach in the darkroom) by using several printings with different carefully chosen ink colours. When two inks are used, this is called duotone, and tritone and quadtone printing are also used where quality is of the essence.
If you want to print black and white using normal inksets on colour ink jets, then it is probably worth looking at the options in Photoshop for duotone (or tritone/quadtone) printing. The deliberate colours introduced are possibly easier to control than striving for neutrality. Again this is outside the scope of this feature and may be dealt with separately.

Bitmap
We generally use the term bitmap for any image that is represented by data relating to pixels, but originally it meant the simplest representation of an image bitmap, where each byte can be on or off, representing black and white. You may think that this would still be a useful way to show things such as type or line drawings, but in fact it is seldom very satisfactory. To get lines or characters to look sharp on screen or in print, we need to use a technique known as anti-aliasing, making use of grey tones.
The left hand of these two letter 'a's is using anti-aliasing, the right in a simple black and white image. You can see the difference in the images from the x10 enlargements shown in the smaller images.



Part 5: Circles & Spheres

Munsell
One of the first systematic ways to organise colours was proposed by an American artist, Albert Henry Munsell around the turn of the 20th century. Munsell was also the inventor of the daylight photometer. Building on the work of O N Rood, published in 1879, he started with a colour wheel with the different hues arranged around a circle, going through the spectrum from red to blue and then continuing with the purples formed from mixtures of blue and red to complete the circle.

Decimal Color
These hues were named in terms of five principal colours - red, yellow, green blue and purple. Midway between each of these principals he situated 5 intermediate colours, yellow-red, green-yellow, blue-green, purple-blue and red-purple. Munsell's system was thus a decimal system of colour, with further subdivisions of each of the ten colours into ten degrees around the circle.

Hue, Value and Chroma
As well as its hue, Munsell also characterised colours by how light or dark they were - which he called 'value' and also by the colour intensity which he called chroma. The value depends on the amount of light the colour reflected and in other systems is called brightness or lightness. Chroma is also known as saturation.
Munsell's work showed clearly that some colours - particularly red, blues and purples - had a much wider range of chroma than others such as yellows. Also, at full saturation, colours such as yellow are considerably brighter than blues or purples. He had hoped for a neat colour sphere by placing chroma values of each hue radiating from the centre of the circle, with different values represented on a vertical axis, but the final result was a rather distorted figure, sometimes called a colour tree rather than a colour sphere.
Munsell's system is based on how we perceive colours, and variations on it have appeared in many books and courses on painting. Munsell colour atlases and charts, available on different surfaces, are still an important tool in getting printed material the correct colour, and also in testing for correct colour vision.

Colour Pickers
Colour pickers in computer paint programs often make use of systems loosely based on Munsell's work, using HSB (Hue, Saturation and Brightness) or HSL (Hue, Saturation and Lightness) models, to choose colours for use in RGB or CMYK colour spaces.

Pantone
Another widely used system of predefined colours is the Pantone Matching System, the most widely used method for accurate matching of spot colour printing. This was largely developed as a practical system in the 1960s.
Although neither of these methods is actually used as the basis of a colour space in computer colour models, they are often the basis of the colour pickers by which you can select available colours. Methods that are based on comparing with printed samples will only give accurate results using those printed samples, and these cannot be accurately represented on a monitor using RGB.

A 2 Z of design & advertising

A
Ad copy: The words (printed or spoken) contained in an advertisement.

Advertising: The general public usually uses this term for any and all kinds of promotion. Professionals, however, only consider paid communications, that are placed in mass media (newspapers, magazines, radio, TV, etc.) to be “advertising.” Other activities are excluded (e.g., sales promotion, direct selling, public relations, etc.).

Advertising allowance: Money paid by manufacturers to distributors or retailers to encourage them advertise a specific product to customers.

Advertorial: An advertisement that is designed to simulate the appearance of a news article or editorial.

Arbitron: A rating service that publishes reports that measure radio and television listening / viewing audiences.

Audience: The number of persons or households exposed to a medium – regardless of whether they saw or heard the ad message. The audience size is based “paid subscribers” (print), “listeners” (radio),”viewers” (TV), or “site visits” (Internet).

Artwork: The visual (photos or drawings) elements of an ad or collatoral printed piece – not including the typeset text.

Average audience: In print media, the number of readers who saw an average issue of a publication.

B
Banner: An ad on a Web page, usually a rectangular shape at top or bottom of the page, that contains the advertiser’s message.

Billboard: Two very different meanings: 1) an announcement or sponsor identification promoting a radio or TV show; 2) an outdoor sign or poster.

Bleed: In print media, extending the color, photo, or artwork beyond the normal margin, to the edge of the page.

Blueline: Name given to a final proof from a printer. The proof includes a “blue color” line drawn on a mechanical to indicate where the finished page will be trimmed.

Body copy: The text of an ad or printed piece – excluding the headline, sub-headlines, or logotype.

Boldface: Type with thick, heavy lines

Brand name: Name that distinguishes a company’s products or services from its competitors.

Broadsheet: Describes a standard size (13” wide x 21” high) newspaper.

Business-to-Business Advertising: A program of advertising and collateral materials directed at business, rather than consumer, audiences.

Button: A Web page ad, smaller than a banner -- often linked to advertiser’s own Web site.

Buying service: A consultant who provides the service of planning and buying media space (print) and time (broadcast). Usually provides no creative or production services.

b/w: Abbreviation for “black & white” – an ad or other collatoral material printed black ink on white background.


C
Camera-ready art: Artwork that is finished to the point of being ready to be turned-over to the printer or publisher.

Caption: The text positioned adjacent to a photo or illustration.

Card rate: Usually the highest rate charged by broadcast media or print publications. Published as a printed rate card, or price sheet, with any discounts calculated as a percent of the “card rate.”

Circulation: In print media, the number of publication copies sold or distributed.

Classified advertising: Generally found in newspapers and selected magazines, and organized according to specific product/service categories.

Click: The number of times users click on an online banner or button ad.

Click through (also: “ad click rate”): Percentage of ad views that result in an ad “click” (above). Results from a user clicking from the host Web site through to the advertiser’s Web site.

Closing date: Deadline set by print and broadcast media for the receipt of advertising material scheduled to appear in a specific publication issue, or broadcast time slot.

Collateral materials: Not considered “mass” advertising media. Rather, these materials are considered to be supportive of the sales effort, usually distributed at the point-of-sale (e.g., sales brochures, catalogs, specification sheets, mailing pieces, etc.).

Color proof: An early (before final approval for press run) color print of the printed piece or ad, produced to judge final appearance.

Color separation: The basis for a full-color ad or printed piece – generated by printing 4 separate colors: yellow, cyan, magenta, and black. The actual “separation” consists of four separate screens – one for each color.

Column inch: A unit of measure used by publications to describe the total space purchased to display an ad. Ad space is defined by two dimensions: Columns (equal to the width that a publication sets its text) timesInches (depth). For example, if the ad space required is 20 column inches, then this allows for an ad that spans 5 columns wide by by 4 inches deep. A twenty column-inch purchase may also accommodate an ad that is two columns wide by ten inches deep. Nonetheless, in either example, the cost of the purchased ad space is the same (Price Per Column Inch x 20)

Comprehensive layout (also: “Comp”): A rough layout of any graphic project (e.g., ad, collateral piece, packaging, etc.) The layout should be detailed enough to approximate how the finished project will appear.

Concept & design: Three elements combine to develop any advertising communications piece – the creative idea; the visual (photos and artwork); and the words (written or spoken). The process can be accomplished by a single person, but usually involves a team consisting of client, layout artist, and copywriter.

Consumer advertising: Promotion directed at the buyer who will utilize the product or service for personal, rather than business, use.

Continuous-tone art: A photo or other art that enables printing to be produced with blended gradations of color, or light and dark shades of gray.

Cooperative program (also: “Co-op”): An arrangement between two or more parties (manufacturer- distributor-retailer) so that advertising and/or promotion costs are shared by them.

Copy: Written words in printed materials; or, words to be spoken in broadcast scripts.

Copywriting: The creative process of writing text for advertising and collateral Materials, or scripts for broadcast advertisiments

Copy editing: The process of approving and changing written text or scripts before final approval.

Cost per (rating) point (CPP): The cost of buying broadcast media measured as one percent (a “point”) of a specific targeted audience.

Cost per thousand (CPM): When buying a specific advertising or promotion medium, the results are expressed in terms of the cost to reach 1,000 people.

Crop: The act of marking specific areas (edges) of a photo or illustration that are not to be shown.

Crop marks: The actual marks that indicate specific portions of the photo or illustration that are to be deleted.

D
Double-truck: A newspaper advertisement consisting of two facing pages.

F
Final file format: The specified type of software that artwork files must be compatible with when delivering materials to a publication or printer. Common fomats include Adobe Acrobat (.pdf), Adobe Illustrator, Adobe Photshop, Adobe Indesign, Quark Express.

File archiving: An electronic copy of the artwork as it was composed, along with all art elements that were incorporated into the layout (images, logos, etc). These elements are stored usually on removable electronic media (CD, zip, etc) so they can be accessed in the future if needed.

Four color process: The printing technology that involves use of four ink colors [black, red (magenta), blue (cyan), and yellow] to produce a printed image that matches the full color of the original artwork.

H
Hit: The total of every component of a Web site that is viewed by a visitor (e.g., a Web site with 10 graphics, and the viewer clicks them all, results in a total of 11 “hits.”)

HTML: Hyper Text Mark-up Language. A basic programming convention that contains instructions for properly displaying web pages in Internet browser windows.

Hyperlink: A method of associating words on a web page with other web pages or email addresses. Hypertext is generally indicated by a different color, and often by an underline, to distinguish it from regular text and allow the computer cursor to be placed over the distinguishing words and “clicked” in order to launch the associated page or email address.

K
Kerning: The amount of space separating each character in a word.

LLayout: The physical appearance of ad materials as it appears on a sheet.

Leading: The amount of space separating lines of text.

Logotype (logo): The unique design created for a corporate symbol, brand name, publication title, etc. Usually presented in a typeface or special hand-lettering format. Logos are usually registered as trademarks for exclusive use by the originating client.

M
Marketing mix: The total elements of a firm’s product / service marketing efforts that influence sales results. (e.g., product styling or service features, pricing, packaging, advertising, sales promotion, public relations, service, distribution etc.)

Mechanical: A camera-ready paste-up that combines artwork and type elements onto an art board or computer disk. Generally speaking, it is your ad composed in a format that is acceptable for publishing.

P
Packaging design: The layout of graphic and text elements of a product enclosure or container. A good package design will also act as a dynamic point-of-purchase display at the distribution point.

PDF: Portable Document File. Any text or image file saved in a format that is universally readable using Adobe Acrobat software on any operating system. It offers the advantages of capturing images and fonts so these elements do not need to be separately imported in order to view the document in its original layout. This makes file sharing easier across different computers, often including camera-ready files for publication.

P-O-P: Abbreviation for “point-of-purchase.” Generally refers to display or packaging material that is aimed at encouraging sales by consumers at retail outlets.

RRack Brochure: A sales collateral piece designed to fit in a standard #10 business envelop, often pre-folded and consisting of two, three, or four panels.

Rate card: A printed price schedule that specifies the base costs (usually highest rates) for advertising in a print or broadcast media.

Royalty Free: Photographic images priced without limits to how often or where they can be used by the purchaser. Compare to, Rights Managed.

Rights Managed: Photographic images, which are priced according to a particular use, the cost of which may vary depending on the size it appears, the number of different advertising vehicles it is used in, the duration of its appearance, and the industry it is used to promote. Compare to Royalty Free.

SSales promotion: Usually includes a wide variety of supplemental sales-generating activities. Usually a two-level marketing effort directed at both “dealer” and “consumer” groups. Includes a wide variety of supplementary sales activities including personal selling, advertising, collateral printed materials, public relations, etc.

Snipe: A small band of short text, usually placed in the corner of an ad, designed to call attention to a new or unique feature of the product or service being advertised.

Spot: A commercial that is scheduled within a radio or TV program.

Spot color: The use of an extra color for emphasis – generally, in addition to a basic black-and-white layout.

Stock image: A photo or illustration that is available for purchase and use by any advertiser. A wide variety of stock images are available, at prices far lower than the cost of creating original artwork. The nominal purchase cost is usually based on the specific medium to be used (collateral, print advertising, etc.) and the anticipated size of the target audience.

T
Tabloid: A newspaper that is roughly half the size (11” wide x 14” high) of a standard newspaper page.

Typeface: The size or style of a specifically designed alphabet. Type can express the tone of your communication – humorous, serious, attention-getting, etc. There are hundreds of typeface choices.

Typographical error (also: “Typo”): A mistake in printing, typing or writing.Typography: The art and technique of selecting a specific typeface design (size, weight, and style) for maximum readability and appropriate aesthetic appearance.

W
Web site: Promotional pages saved in file formats that are readable by Internet browser software and posted on servers that are connected to the World Wide Web. Establishing a Web site involves 1) registering a domain name (eg, admedic.com), 2) creating effective site content navigation and page construction, and 3) establishing a host or Internet Service Provider (ISP) to store web pages on a server that is associated with the domain name.

Y
Yellow Pages: A section of a telephone directory that lists ads for businesses, services, and products.


Source: www.admedic.com

some of my CLICKS