Index terms: paper, whiteness, reflection spectra, cluster analysis
Abstract: Standards imposed on the quality of printed products require precise compliance with the whiteness of the printed material to certain standards. There are a lot of offers from various manufacturers on the market of printed papers and cardboards, which differ significantly in manufacturing technology. A spectrophotometer can be used for operational input control of paper whiteness in printing houses. The obtained spectra must be analyzed and systematized. The whiteness of printing paper and cardboard is the most important property for the formation of high-quality multi-color half-tone image. At the same time, the overwhelming majority of printing paper grades have a certain shade due to the manufacturing technology, as well as the presence of various mineral components introduced at different stages of production. To date, no systematic information has been found in the literary sources on the influence of the composition of printing paper and coating layers on its reflectance spectra, as well as on the ranking of printed paper according to their optical properties. In this paper, to systematize information on the optical properties of coated printing papers, the reflection spectra of 254 samples were analyzed. Using cluster analysis methods, three statistically different clusters were ranked in the paper samples, ranking the samples according to their reflection spectra. As a result of the study, it was found that the CIE Lab color coordinates calculated on the basis of spectral data are a convenient criterion for clustering samples according to their spectral characteristics. The selected three clusters demonstrated a significantly different character of the average reflection spectra. Further correlation analysis revealed some regularities in the mutual influence of monochromatic reflection coefficients and color coordinates within clusters.