Technology companies have been seeking to provide a solution for this missing piece of the workflow in the form of color management systems, which are essentially systems for managing color reproduction between devices. These systems are the crucial piece of technology that holds the potential for both consumers and professionals to achieve the same level of quality and consistency as experienced craftsman.

The goal of a color management system is to provide color consistency and predictability between devices. The goal of color matching is a panacea—it is not possible to obtain an exact match between different devices (more on that in a few minutes). Color management system provide predictability and consistency, reducing the guesswork and cost involved in reproducing images and artwork. This important distinction is a key in making color management work.

What is a color management system?

Color management systems take various forms, but at their most basic, they consist of three core elements: an engine that transforms color contained in images and documents, devices profiles that represent the color capabilities of a printer, monitor, scanner, or camera, and an interface that allows users or applications to utilize and manage the capabilities of the system.

Ideally, color management systems are implemented at the operating system level—providing broad, standardized color management to users and applications of a particular platform. This is accomplished by providing an open architecture that supports different devices and applications, industry standards, and a powerful method for exchanging color information between input devices, displays, applications, and output devices. Fortunately, the two mainstream operating systems have both implemented system level color management systems, including support for industry standards. Many application vendors have also implemented color management, requiring users to carefully develop color managed workflows that suit their needs.

Why Is Color Management Needed?

If I only had a dime for every time I heard “Why don’t the colors on the monitor match the colors that come out of the printer?” or “How come digital prints look different from scanned originals?” There are many reasons, based in complex color science, why the appearance of a color image is difficult to predict on different devices. In many ways, communication in color has problems similar to those with communication in languages. Each device is like a person, speaking his or her own language. When one person that speaks French attempts to communicate with another person that speaks Japanese, there is a breakdown, and the message is not communicated accurately, if at all. What’s needed is an interpreter that is capable of interpreting the language as well as the dialect, to ensure that the message is properly communicated. The same holds true for color reproduction.

There are a few basic concepts to understand when working with color. The first is how devices reproduce color, and the second is how desktop devices communicate in color. Each device that is capable of reproducing color has a range of colors that it can accurately reproduce, better known as the gamut. Each device, such as a display or printer, has a unique gamut which is dictated by the characteristics of that device—the type of technology it uses to print, the type of phosphors, etc. Even devices of the same type, say two photo printers, will have different gamuts. A device’s gamut is a subset of a larger standard area of color, known as a color space. Different types of devices work in different color spaces. Monitors and cameras work in RGB, while printers and scanners come in both RGB and CMYK flavors. Printers such as Epson’s Photo Stylus line are RGB devices that use 6 or 7 colors of ink—CMYK inks. It’s easy to become confused! What’s important to understand is that device color spaces are different, which helps to understand why it is difficult to match colors between different devices.

The gamuts of desktop devices, such as displays and printers, are relatively small when compared with the visible spectrum of light. Color scientists use mathematical models that represent color in different ways. For the purposes of color management, models that represent the visible spectrum are used, as they easily contain all of the colors that an imaging device can capture or reproduce. The following illustration shows a color model common to publishing, CIE LaB, with the gamuts of a scanner, monitor, and printer overlaid.

CIE Color Space

The color spaces typically used by devices are made up of the additive or subtractive primary colors. The additive colors, red, green, and blue (commonly referred to as RGB), are used by all color displays and many scanners and cameras, and some printers. The subtractive primaries, cyan, magenta, and yellow, with the addition of black (commonly referred to as CMYK), make up the color space used by output devices, as well as some scanners. The two color spaces also have many areas that do not overlap. This illustrates two key points: different devices reproduce color differently, and the human eye can see far more colors than today’s devices can reproduce.

This is easily illustrated by a familiar problem:

The colors seen on a display are typically much brighter and more saturated than the color that comes out of the printer. The reason for this is likely that the color on the display is in gamut for the display, but is not in gamut for the printer. The printer simply cannot reproduce all of the colors contained in the image. In this case, the color is said to be device dependent–the desired appearance of color depends on its being reproduced on a particular device. Several different factors determine which color is substituted for the original color, better known as gamut mapping.

The second challenge exists in the method that desktop devices communicate with each other. It’s typical to have a camera, display, and printer from three different manufacturers, as well as image manipulation and page layout software from different software vendors. But none of these devices communicate in the same language of color, and many handle color differently—much like the example of human speech.

Color management systems link these disparate pieces together, and attempt to provide a consistent way for devices and applications to interpret color data.

The Color Management Wave

Early color management vendors developed the basic approach to address the problem of device-dependent color. The fundamental solution was to use profiles for devices, and a color transformation engine to convert colors from one color space to another. Device profiles capture the gamut of a device—essentially a color dictionary that can translate between a device’s color language. In addition to device profiles, device-independent color spaces were utilized to provide greater flexibility. These are color spaces where the definition of a color is not dependent on any particular device. The idea is to use color spaces that represent the entire range of visible colors as translation spaces between different device spaces. This means that any color that is shown on a monitor is within the gamut of this neutral color space. In 1931, the CIE (Commission Internationale d’Eclirage) established standards for a series of color spaces that represented the visible spectrum–60 years before the arrival of desktop color! The CIE color spaces form the foundation of device-independent color for color management. Many of these spaces, such as CIE XYZ and CIE Lab, are widely used in color management systems today. These color spaces, along with several other pieces that will be described in the following sections, together form a system for managing and matching colors.

Device Characterization and Calibration

Early color management system used proprietary profiles. Companies such as Agfa, Kodak, EFI, and others had developed systems that were not compatible with each other. This made them very difficult to use in an open environment, such as a Mac running Photoshop. Fortunately the International Color Consortium was founded to establish color management standards, the first of which was the ICC profile. ICC profiles are based on a well-defined standard, and are now supported by virtually all vendors of color imaging hardware and software, as well as Microsoft and Apple.

Profiles are basically dictionaries that contain data on a specific device’s color information, including its gamut, color space, colorants, and modes of operation. The process of creating profiles is known as device characterization. Device characterization is typically performed with highly sensitive color measurement devices. The resulting measurements are input to a software package that uses several complex algorithms, to create a profile. Today there are numerous color measurement devices and software packages for creating profiles (device characterization and related products are discussed in more detail in the forthcoming color management e-booklet)

Most color devices come with profiles, so why would you want to create your own? It is important to recognize that these profiles represent the device in its factory condition. In reality, devices change as a result of time, materials, and environmental conditions, resulting in inconsistencies from the stock profile. Creating a custom profile ensures greater accuracy, and hopefully greater color predictability.

Several profiles have been created as industry-standard color space profiles. Microsoft and HP worked to develop sRGB, a generic RGB color space designed to represent the average PC owner’s monitor. Adobe has created AdobeRGB, similar to sRGB but tailored to the publishing professionals. Color management experts, such as Bruce Fraser have created custom profiles, like BruceRGB, to address specific challenges and workflows. In many cases, these profiles can improve the predictability and reliability in color management, particularly for digital cameras. Today, many digital camera manufacturers are using these color spaces in their cameras.

A process known as calibration is often used to ensure that a device is performing to factory specifications. This process is often simpler and faster than device characterization, and can be performed on a regular basis to ensure accuracy. Some devices, such as some of the HP printing devices, are self-calibrating, greatly reducing the need to create custom profiles. Many of today’s characterization tools perform both calibration and characterization simultaneously. Monitors, such as Sony’s new calibrated display, not only include calibration and characterization tools, but monitoring tools to alert users when they need to re-calibrate.

How does it all work?

In theory, it’s straightforward. Profiles are used by a color transformation engine, more commonly known as a color matching method (CMM). The CMM translates data from one device’s colors to another, via an independent color space. The CMM receives the necessary information from the profiles, so that it can accurately transform a color from one device to another. The result is hopefully color that is predictable from device to device. It is not possible to have perfect color matches between devices due to differences in each device’s gamut. For example, many of the deep, saturated blues and greens that appear on a display cannot be reproduced by printers using the CMYK ink set. In this instance, the CMM must perform gamut mapping, a process by which the next closest reproducible color is selected. Most matching systems offer several gamut-mapping methods, or rendering intents. Because the use of color varies from business graphics to photographic reproduction, the rendering intent of a color must be specified to produce the best possible results.

One of the fundamental problems that prevented widespread adoption of early color management systems was the fact that each was implemented using a different architecture. In order to perform color-matching functions, an application manufacturer would have to select one system and then make specific calls to it. However, because there was no common color management framework for applications to use, each application had to use a unique system, with no compatibility between profiles and no consistency among the results.

Because no single system was widely adopted, all of the systems failed to provide a satisfactory solution. From a user’s perspective, there was no guarantee that peripherals and applications would all work together to provide a complete work flow with consistent results. And because each system was proprietary, users could not exchange files, like profiles, with users of different systems.

Platform color management system

To address many of the issues surrounding color use, Apple Computer introduced
ColorSync in the early 1990’s. The goal of ColorSync was to provide a common
architecture for color management systems. Microsoft followed suit within a couple of years. Both vendors support the ICC profile format, enabling device profiles to be platform independent.

Color management system such as ColorSync and Microsoft’s ICM provide interfaces that software developers can use in developing their application. This means much less work for application vendors–they need only call one interface for all color management functions. Digital photography applications such as BreezeBrowser and Thumbs+ utilize color management by calling the color management system in the operating system.

Today’s vendors have done an excellent job of implementing color management system. Most popular imaging applications support a variety of different color management features. Leading vendors such as Adobe have implemented color management in the same manner between most of their applications, ensuring that color data is consistent as users move from one application to another.

There are many different steps in a color managed workflow. They key to successfully implementing one is consistency. It is easy to create overlapping steps in your workflow which result in double color matching.

One benefit of color management that is widely supported is the ability to simulate the color output of one device on another, known as proofing. Soft proofing simulates the output of a printer on a display, while output device proofing simulates one output device on another. Many applications also offer a gamut alarm, which alerts users when a color can’t be accurately reproduced on a particular device. Proofing can save you time and money!

Color management has reached a level of adoption such that it can save time and money, and can be easily implemented. Like digital photography, a great deal of money can be spent on color management products. Outback Photo contains many useful resources for color management, including device reviews and workflows, that hopefully will make your experiences with color management more pleasant.

About the author

Joshua Weisberg has extensive experience in the computer and publishing industry, with specific knowledge of color management, color imaging and digital photography, and is the co-author of the book The GATF Practical Guide to Color Management, a comprehensive guide and resource for color management.