Other Free Encyclopedias » Online Encyclopedia » Encyclopedia - Featured Articles » Contributed Topics from A-E

Color Management in the Future

icc image device workflows

JEAN-PIERRE VAN DE CAPELLE, Ph.D.
Xerox Corporation

Today’s color management in the commercial printing community is dominated by two types of workflows: ICC workflows 1 and more traditional, “device CMYK” workflows. In many cases the device CMYK workflows use ICC color management in the prepress arena to convert digital RGB images to CMYK. Consequently, one could argue that ICC workflows actually dominate the space. For many applications, ICC-based color management fulfills most of the needs of the typical CMYK printing used by a large number of commercial printers, irrespective of how the print-stream data is formatted or which color space is used within that print stream. As such, one should realize that the incremental economic value of different, future color management systems will be relatively small for this market. One can argue that most of the needs of the commercial printer were already fulfilled with the ICC version 2 or version 3 specifications, and that the incremental value of the version 4 specification is rather small. This version provided some clarification as well as rendering intents for input and display devices. This does not mean that today’s color management has no problems. The remaining problems are often of a very fundamental nature, such as device color stability, significantly different device color gamut, or the influence of the illuminant on the color rendition of the printed material, which can include metamerism and fluorescence. Metamerism is a psychophysical phenomenon commonly defined as a situation in which two samples of a color match under one illumination condition, but fail to match under another condition.

For other applications, industries, and markets such as packaging, security, decoration, silk, ceramics, automotive, photo-finishing, medical imaging, and biological imaging, ICC color management is not sufficient to meet daily needs. Some of the requirements of those applications may also apply to smaller segments of commercial printing; for instance when metallic or spot inks are used, or when there is a need for using automated corrections in the workflow, such as “red eye” removal or contrast adjustments.

In addition to this, color management is almost unilaterally perceived as complex and tedious, and in many cases it is actually expected to work without too much effort. This desired ease-of-use state has not yet been achieved.

In 1993, when the ICC was founded, the agreed-upon principle for color management implied that color rendering is predominantly determined by the data in the profiles and, to a far smaller degree, by the color transformation engine, known as the Color Management Module (CMM). This is often referred to as a “dumb” CMM, and “smart” profiles. One of the main reasons for this choice was the compute-intensive nature of a “smart” CMM. One of the problems with this approach is that the “intelligent” profiles from different vendors may not work together as well as desired. Given increasingly faster CPUs, this paradigm could be reversed and profiles could be made "dumb"—in other words, to only reflect the device response—and the CMM could be made smart, determining how the color rendering will be done and incorporating the effects of different device color gamuts and viewing conditions. In theory it would even be possible to adapt the color-gamut mapping to the gamut of the image in its “original” color space and the available gamut of the reproduction device to represent this image. This would possibly lead to better reproductions of each individual image. It would also mean that profiles provided by different vendors would be more compatible with one another.

As such there are opportunities for improved color management technologies beyond ICC in different areas of the supply chain, from design to print production and finishing. The landscape implies that there are different futures for color management technology, depending on the type of application and market.

Within its new Vista OS, Microsoft, with its new Windows Color Management System (WCS), has chosen the “smart CMM, dumb profile” paradigm for color management, while maintaining support for the traditional ICC workflows 2. From the disclosures provided by Microsoft, WCS does not yet provide image-dependent color management. It is worth noting that the procedures used with high-end reprographic scanners in the past incorporated image-dependent and often localized processing techniques, such as black-point and white-point settings, cast removal, tone curve corrections, selective color correction, etc. These procedures were very much dependent on human interaction and judgment, and they usually led to high-quality reproductions at a considerable price, because color was understood by few and the scanner operators were therefore well paid. The incremental economic value of WCS over the current ICC-based workflow remains to be established.

Generally there are a couple of larger technological areas that constitute the incremental fundamentals needed for different applications:

  1. Special materials and application, such as metallic substrates and multi-ink or special (non-CMYK) ink systems, and also medical or biological samples.
  2. Automated color processing for preferential rendering, automatic tagging, and retrieval of images in searchable image databases.

Additionally there is the underlying assumption that the ease of use for any color management system should improve.

An example of technologies developed in the first main area includes spectral characterization of inks in multi-ink printing processes 3. Examples of technologies being developed in the second area include “Smart CMMs”, such as WCS, and Automatic Image Enhancement for photofinishing applications .

Some people may argue that some of these technologies, such as segmentation and image content recognition (ICR), are not strictly color management, but should be considered complementary technologies. Segmentation, for instance, can be used in object-oriented rendering, and ICR can be used for preferred pictorial rendering.

Color Measurement - Introduction, Historical Perspective, Definitions and Terminology, Components of a Spectrophotometer, Light Source, Detector, Dispersing Element [next] [back] Color Management - Introduction, ICC architecture, ICC profile, Profile Connection Space (PCS), Module (CMM)

User Comments

Your email address will be altered so spam harvesting bots can't read it easily.
Hide my email completely instead?

Cancel or