The Grand Canyon of Color: How to Talk About Color Across the Creative / Technical Divide

John Seymour, John the Math Guy, LLC; Maggie Maggio, Smashing Color

Joint presentation on how to talk about color between creative & technical teams, 45-minute presentation.

Update to Evaluation of Reference Materials for Standardization of Spectrophotometers

John Seymour, John the Math Guy, LLC

Statement of problem to be solved

Reference materials are physical standards that can be used to improve the color measurement process in several ways.

Use case 1: Reference materials can be used to verify that a color measurement device has not changed. The set of physical standards is measured initially, and then again at regular intervals. If the measured values do not change appreciably from the initial measured values, then it is likely that the measurement device continues to measure faithfully.

This technique can be robust provided the set of reference materials is physically stable, that is, resistant to scratches and fading. Ceramic color tiles are popular as a reference material for this reason.

For this use case, there is not a strong need for a broad set of colors; it has been suggested [Berns and Reniff, 1997] that a single cyan tile may be adequate.

Use case 2: Reference materials can be used to quantify the agreement between two color measurement devices. This quantification may advise the user as to the suitability of mixing the two devices in an application.

The requirements of the reference materials for this second case are somewhat different than for the first case. The physical stability of the reference materials is less critical, especially if the two instruments are to be compared sided-by-side and at the same time. But the need for a wide range of colors (or more accurately, spectra) is essential to accurately assess the agreement of the instruments.

A less obvious requirement is that the reference materials should be similar in nature to the samples that will be measured. It has been found that differences in the illumination and acceptance angles between two different models can be a substantial source of disagreement. In addition, differences in aperture may create a similar problem due to the lateral diffusion of light within certain of the ceramic tiles [Seymour, 2014, ISO/TS 23031].

Thus, instrument agreement based on measurements of, for example, high-gloss ceramic reference materials may not be representative of disagreement that will be seen when two color measurement devices are measuring samples of print.

Use case 3: Reference materials can be used to standardize one spectrophotometer to another [Robertson 1986, Berns and Reniff 1997, Rich and Martin 1999, Van Aken 2000, 2003, 2006, Chung et al. 2002, Rich 2004, Nussbaum et al. 2011, Seymour 2013]. This is the practice of using a set of reference materials to calibrate a transform which converts spectral reflectance values from one spectrophotometer so as to more closely agree with another spectrophotometer. The most common technique for standardization involves quantifying wavelength shifts.

This approach for standardization relies on moderately strong derivatives of reflectance versus wavelength at all wavelengths. It has been pointed out [Seymour 2013] that the Lucideon Colour Standards Series II (CCSII) (AKA BCRA tiles) lack such strong derivatives in certain regions of the visible spectrum, and hence are less than ideal for standardizing one spectral color measurement device to another.

Thus, we have a set of ceramic reference materials which is in common use for standardization of spectrophotometers in the graphics arts, but which has two undesirable properties. The ceramic tiles are considerably glossier than printed samples, and they are less than ideal in their ability to calibrate at all wavelengths.

Evaluation of a possible solution

The ISO technical specification ISO/TS 23031 recommends the ChromaCheckerTM Instrument Inspector target as a set of reference materials for comparison between different models of color measuring device. It further suggests that this set may be used to diagnose causes for disagreement. It stops just short of suggesting ChromaCheckerTM as a set of reference materials for standardization of one color measurement device to another.

There are two physical reasons to prefer this set of reference materials. First, the gloss of the ChromaCheckerTM is within the range of print, whereas ceramic tiles are generally considerably glossier. Second, the patches are a thin film, so lateral diffusion is minimized, or is at least similar to what may be found in graphic arts samples. However, the spectral suitability of ChromaCheckerTM Instrument Inspector target has yet (to the knowledge of this author) to be investigated.

Another candidate for a set of reference materials for standardization of color measurement devices is a set of reference materials which Lucideon has developed specifically for the print industry. Since these are ceramic tiles, they may not match the gloss characteristics of print. As such, they may only be suitable for standardization between spectrophotometers with the same make and model. Again, this author is unaware of any published assessment of this set of ceramic standards for the purpose of spectrophotometer standardization.

This paper fills that need by repeating some of the analysis that was performed on the set of the ceramic reference materials in Seymour 2013. For comparison, a similar analysis is performed on a set of paint samples.

Bibliography

  • R.S. Berns, R. S. and L. Reniff, An Abridged Technique to Diagnose Spectrophotometric Errors, Color Research and Application, 22(1), 51(10) (1997)
  • ISO/TS 23031, Graphic technology — Assessment and validation of the performance of spectrocolorimeters and spectrodensitometers, ISO TC 130,
  • Rich, D. C., and D. Martin, Improved model for improving the inter-instrument agreement ofspectrocolorimeters, Analytica Chemica Acta, 380, (2-3), 263-276, (1999).
  • Seymour, John, Evaluation of Reference Materials for Standardization of Spectrophotometers, TAGA 2013
  • Seymour, John, Assessment of the Sources for Disagreement Between Two Spectrophotometers, TAGA 2014
  • Van Aken, Harold, and Ronald Anderson, Method for maintaining uniformity among color measuring instruments, US patent 6,043,894, 2000
  • Van Aken, Harold, and Ronald Anderson, Method for maintaining uniformity among color measuring instruments, US patent 6,559,944, 2003
  • Van Aken, Harold, Shreyance Rai, Mark Lindsay, Richard Knapp, System and method for transforming color measurement data, US Patent 7,116,336, 2006