Colour management is a collection of rules and procedures in digital imagery which govern the control and management of every process involved in the colour transformations which occur during image capture or rendering images between input devices such as a digital camera or scanner and output devices such as our monitors or printers which render our images with ink on a substrate.
To boil it down, colour management is rather like a protocol for the translation of one language into another - say French into English with a person between having the ability to handle the translation between the two - except that instead of languages the subject for translation is a product of light. An example is image data captured by a digital camera which must be translated so that it can be shown by, that is, output on a screen display or printed with ink on paper.
So, you might ask why is a translation necessary and what are we translating?
First, the primary components of light are Red, Green and Blue (RGB). Various mixtures of red, green and blue light reveal all the colours that we can see and the combination of all three in equal amounts produces white light. When we make a photograph with our digital camera, it captures the light which is reflected from or transmitted by objects in the scene. This light is broken down in our cameras to RGB data. The camera “sees” the same scene that we see but it is not able to collect all of the colours that we see nor in exactly the same way that we see them. Many colours may differ from the actual scene significantly.
Secondly, using light, the monitor on which we would display that image works by recombining the RGB data from the camera back into the form of an image. It is also limited by the number of colours it can display which is not as much as we can see nor as much as a camera can capture. The colours it can display may also be quite different from the those which the camera captured and the actual colours which we see in the scene.
Finally, to complicate things further, when we print using standard inks the primary colours for printing are Cyan, Magenta and Yellow instead of Red, Green and Blue. This is true because the inks which are transparent act as filters. When light hits printed patches of cyan, magenta or yellow ink, the red green and blue components of the white light which hits them are filtered out and the remaining complementary colours cyan, magenta and yellow are reflected back to our eyes. When these inks are printed in various mixtures we see different colours by the light which is reflected from the printed piece. But once again the range of colour which is available by printing is far smaller than the range of colour which is captured by our camera or scanner or displayed by our monitor or certainly by the amount of colour that we are able to see. Printed colours will also be considerably different from those which were actually seen by the human viewing the scene or those which were captured by the camera or displayed by the monitor. Bear with me. This will all make sense shortly. Sounds like a real mess, right?
Well, to sum all this up, the range of colour which a device can capture or display is referred to as its gamut. As mentioned, the input devices, camera and scanner, and output devices, monitor and printer, respectively all perceive or output colour very differently from one another and from the original scene. They all either capture or produce different sized colour gamuts - some larger and others smaller and many of the colours which they capture or output differ from the actual colours in the scene. So, if you were to take the digital image file from the camera and simply display it on a monitor or print it without the support of colour management to facilitate accurate translations of colour between the devices, the image on the monitor or in print would appear very clearly wrong as compared to the original scene. Also, the image displayed by the monitor would be considerably different from the same image in print.
For example a vibrant red in the scene may be interpreted as orange by one device or burgundy by another and brown by yet another. By the time we view the output of the digital file from the camera on a monitor and in print and we see that orange, red and brown how do we know which of all the different interpretations is true, if any at all?
We must be able to capture images and have our monitors display them and printers print them accurately so that in the end the output looks like the subject of the photographs and consistency is maintained across all the devices in terms of colour and tonality.
When colour management is used properly it ensures colour accuracy during the colour translations which take place as the original photograph is displayed by a calibrated monitor through to print on any given substrate by any type of printer. This feat of technical brilliance is accomplished with the implementation of ICC profiles which are an integral part of colour management.
So here is how it works. ICC profiles are small software modules or data files which provide characterizations or snapshots of devices in terms of exactly how each device interprets colour. It is an encapsulation of the particular colour language of each device. So, for the above examples each device - camera, scanner, monitor and the combination of printer plus ink and a particular paper - would have its own accompanying ICC profile. The system was developed in 1994 by International Colour Consortium (ICC).
As mentioned above, each device perceives or outputs colour in its own way. In order for an ICC profile to encapsulate these characterisations, every colour which each device captures or outputs is assigned a set of numbers or values. Each value set defines the coordinates of that colour within the library of all the visible colour in human vision. The numbers which relate to human vision are based upon studies and experiments performed and systems devised in 1931 by the International Commission on Illumination (CIE) in France.
Since human vision is the common denominator and every colour from every device has a number which defines it relative to human vision, then that device can be characterised in terms of its colour capabilities, its gamut etc. The ICC profiles comprise this information and each is specific to a particular device.
Now, the equivalent to the additional person who provides the translation from French to English mentioned in the early on example above is a software module called the Colour Management Module (CMM) a.k.a. Profile Connection Space (PCS). The CMM is a software algorithm which resides on our computers. Its purpose is to read the ICC profile for each device in a series and remap the translation of the device colours which each profile describes to their equivalents within the gamut of the next device down the line.
This translation is made in order to adjust the differences between the devices and to align their input or output so that the colour they render is consistent and as true as possible to the original scene or original image as it is transferred from one device to the next. The CMM does this by translating the numerical values of the colours which were captured by or output from a device to represent the colours in the scene. This allows the colour input or output to travel through the processes from camera or scanner to monitor and printer with consistent colour which simulates, as closely as possible, the original scene as it was seen. All the devices will be singing from the same song sheet, as it were.
ICC Profiles are specific to each device. One is made for our camera, another for the scanner, another for the monitor and another for the printer plus ink on paper. With respect to printing, ICC profiles are specific not only to the printer itself but to each combination of ink and substrate. So, if an image is to be printed by a particular printer with OEM ink on Hahnemuhle Photo Rag Baryta, a specific ICC profile would be required in order to define exactly how that particular printer prints with those inks on that specific paper at that time.
A type of "profile" is also required for our images. But, these designations actually represent the colour working spaces in which our images exist in order that their colour can be defined and translated based again upon the common denominator of human vision in order to achieve an optimal result. This topic is for another article. In the meanwhile, I hope this has been helpful.