Specifications

Submitted to Studies in Conservation, March 2006
9
research is still required to develop an improved calibration target for imaging paintings. Guiding
principles should include those defined for exemplifying color-order systems [20].
The color management at NGA was very good. The listed results are typical of a well
color-managed area-array color sensor camera system [21, 22]. The modifications to the Sinar
camera resulted in marked improvement, particularly for the Gamblin target that consisted of
typical artist pigments used in paintings. There was nearly a threefold improvement, though
some of this improvement may be attributed to the Blues calibration target, which better
represents blue artist materials than the ColorChecker SG. An analysis of variance followed by a
multiple comparison test based on Turkey simultaneous confidence intervals using the
studentized range distribution was performed [6] on the CIEDE2000 values to assess statistical
significance. For all the targets, the MCSL-Sinar system was superior to the NGA system at an α
of 0.01.
On average, the MCSL-Sinar system had slightly superior colorimetric performance to
the Quantix-LCTF system. For the ColorChecker DC and Gamblin targets, the MCSL-Sinar
system was superior at an α of 0.01. For the ColorChecker and Blues targets, the two systems
were not significantly different. This is an important result: A six-channel camera had the
average colorimetric accuracy of a 31-channel camera. The maximum and 90
th
percentile errors
for the Quantix-LCTF system were smaller for the Blue Pigments and Gamblin Conservation
Colors targets. The improved average colorimetric performance was a result of the more
complex signal processing combining colorimetric and spectral optimization. The colorimetric
optimization used nonlinear optimization since color differences are nonlinearly related to
incident radiation. This nonlinear optimization was impractical for the Quantix-LCTF since
250,000 independent data points were used to estimate 1,116 coefficients (31 x 36 matrix).
Nonlinear optimization would have been extremely time consuming and convergence to a global
minimum highly problematic. Therefore, the Quantix-LCTF calibration only optimized spectral-
estimation accuracy using linear optimization.
The spectral performance of the Quantix-LCTF and MCSL-Sinar systems are listed in
Table 4. A metameric index was calculated to provide a performance metric in color-difference