Zhaojian Li


In those industries in which materials are colored to close specifications, a means of evaluating the degree of metamerism of colored objects is of considerable importance. Based on Wyszecki's hypothesis and its application to quantifying metamerism as described by Fairman, parameric decomposition is a technique to adjust one spectrum of a parameric match in order to achieve a perfect (metameric) match under a specific illumination and observer condition. This method can be viewed as batch correction using three "colorants" where the color-mixing model is linear in reflectance.

The research in this thesis presented these methods using the basis functions from the CIE color-matching functions (CMFs) as well as alternative basis functions derived from dimensionality reduction techniques such as principal component analysis (PCA) and independent component analysis (ICA) for a pre-defined DuPont spectral dataset and Munsell dataset. 1,152 parameric pairs surrounding 24 color centers were synthesized using an automotive finish paint system and two-constant Kubelka-Munk turbid-media theory. Each parameric pair was corrected to a metameric pair using these various methods. The corrected spectra were compared with the formulated spectra using Kubelka-Munk theory to evaluate the parameric decomposition accuracy in terms of special and general metameric indices. The results showed that the estimated metameric indices from the CMFs-based process primaries presented relatively poor correlation to those from Kubelka-Munk theory. The process primaries from ICA for the Munsell IV dataset showed almost indentical performance in estimation of metameric indices to the process primaries from the PCA for Munsell dataset as well as those from ICA for the DuPont dataset. These three sets of process primaries showed slightly better performance in estimation of metameric indices than the process primaries from PCA for the DuPont dataset.

Library of Congress Subject Headings

Colorimetry--Research; Color vision; Imaging systems--Image quality

Publication Date


Document Type


Student Type


Degree Name

Color Science (MS)

Department, Program, or Center

Chester F. Carlson Center for Imaging Science (COS)


Roy S. Berns

Advisor/Committee Member

Mark D. Fairchild


Physical copy available from RIT's Wallace Library at QC496.3 .L4 2006


RIT – Main Campus