An Overview of Colour Management for Still Images
There is a lot of confusion around the subject of digital colour and the process of ensuring colour consistency for still images. This document provides an introduction to the different methods used to manage digital colour.
Colour Management is one of the most challenging issues facing a digitisation project today and has become something of a 'Holy Grail', stretching the skills, equipment and tempers of many project staff.
Let us first consider what we mean by 'Colour Management' and why it is so important to us. In simple terms it is 'making sure that the colours within an image are as accurate as possible and that our image looks the same where ever it is used'.
It is true, that this might not be important in every case; an image of text is still readable if the page is not quite the right tone, but on the other hand an artistic work or indeed a medical image used for teaching may well be either less effective or misleading if the colour is inaccurate. As digital imaging technology improves, there is a corresponding rise in the demand for higher quality and the expectation of accurate colour in all cases.
Of course colour accuracy has always been a challenge with the reproduction of images, long before we 'went digital' but we are now faced with a further problem. It is impossible to 'see' a digital image, which is only an electronic file made up of '0's and '1's. We can only evaluate the colour of a digital image once it has been output into the 'real world' by a monitor, printer or projector. The colour that we see will be dependent as much upon the 'device' that portrays it as the original image itself.
Device dependent colour
Each and every device in the image workflow will have a different range of colours that it is able to capture or portray, which is called the colour 'gamut' of that device. However the colour gamut of one type of device such as a printer, which uses Cyan, Magenta, Yellow and Black (CMYK) inks is likely to be fundamentally different to the gamut of another such as a monitor which uses Red, Green and Blue (RGB) light.
Although there are many colours that can be represented by both types of device, there are also many others that fall inside the 'gamut' of one but are outside the other.
The colour of a digital image (in 24-bit RGB colour) is described by a value from 0 - 255 for each of its component colours (giving a total of 16.7 million possible colours), however the actual colour of each one of these values is not 'set' but is dependent upon the capability or gamut of the output device. A full '255' value will be brighter and more intense when output on a device with a large gamut than it would be on a device with a small gamut.
Traditionally, colour was managed by standardising the capture and output devices and running test images through the workflow, which could then be compared with the original. Adjustments were made and the test redone. By this iterative process the colour could be reliably controlled within a process called 'closed loop' colour management.
Device independent colour
Although 'closed loop' colour management is very reliable within a known workflow, such as for traditional commercial print, it is hardly appropriate for a modern open workflow, such as the digitisation project, where we may well be using images from a range of capture devices and have no idea how the image will eventually be used. It was therefore necessary to develop an 'open loop' colour management system that fitted the new digital way of working and could manage colour wherever it was being seen and on whatever device.
To provide this, the International Colour Consortium (ICC) developed a colour management system based around the original work of the Commission Internationale de l'Eclairage (CIE) who modelled the approximate gamut of the human eye in the 1930s. The ICC colour management system uses a standardised and known 'colour space' based on human colour vision and then compares all devices within the workflow to a known standard. Each device has its exact 'gamut' measured and then compared against the established standard (colour space). This table of adjustments, which is called a 'device profile' is then used to adjust all images that are captured, printed or viewed by this device. In this way, as long as any device has been measured first (by a process called characterisation) and a profile made, it can be reliably used within the ICC 'open loop' colour management system.
Does it work?
The ICC colour management system has now been widely, though not universally adopted. Following its development by Agfa, Apple, Adobe and Kodak and the incorporation of ICC into Adobe's products the rest of the industry has fallen in line and is supporting this standardised methodology so the future looks secure. As it becomes more accepted, it is hoped that it will become more automatically integrated within the hardware and software and will rely less on the users input and actions to guarantee its effectiveness. At present the system is still reliant upon the end-user being both able and willing to take the necessary steps to establish effective colour management on their own machine. However carefully an image has been digitised and optimised, it may still look completely incorrect on a monitor that has not been calibrated.