Enhancing the Specular Effect of Metallic Color Prints by Reducing the Use of Yellow Ink

The MetallicArt technique is used as a visual security feature to hide patterns in CMY images using an additional silver ink. The patterns are hidden under nonspecular viewing conditions and visible under specular viewing conditions. By varying the amount of silver ink, the authors control the intensity of the specular reflection of the printed pattern. Using a spectral reflection prediction model, they can constrain the maximization of the silver ink (S) in order to keep the color difference between the original CMY color and the fitted CMYS color under a given threshold, ensuring that the pattern is hidden under nonspecular viewing conditions. When the printing device requires the inks to be UV cured, the leafing effect of the silver ink is reduced. This also reduces the specular reflection of the silver ink, which becomes comparable to the specular reflection of the yellow ink. Since both specular reflections are similar, it is difficult to distinguish the embedded patterns under specular viewing conditions when both inks are present. The authors, therefore, propose a new algorithm to embed patterns in CMY images that not only maximizes silver but also minimizes yellow. Experimental results show that the proposed algorithm enhances the visibility of the patterns under specular viewing conditions, allowing the use of MetallicArt when the printing device requires UV curing. VC 2011 Society for Imaging Science and Technology. [DOI: 10.2352/J.ImagingSci.Technol.2011.55.6.060506] INTRODUCTION When light hits a cyan, magenta, and yellow halftone print, it is partly absorbed by the ink layer and partly diffused by the paper substrate, except for the small specular reflection that occurs at the Fresnel interface between air and paper. Although this represents only an approximation of the real phenomena, it has been at the foundation of spectral reflection prediction models and works well in practice. This assumption does not hold for metallic inks. These inks are opaque and reflect most of the light. The incident light does not reach the paper and remains within a relatively narrow cone. The resulting strong specular reflection has been used to hide metallic patterns in normal CMY images, a technique called MetallicArt. Using silver ink, this technique consists in determining the image pixels located within a given pattern and to replace their original CMY coverages by CMYS (cyan, magenta, yellow, and silver) coverages containing as much metallic ink as possible, while preserving the original CIELAB color under nonspecular viewing conditions. The preservation of the original color under nonspecular viewing conditions is achieved using a spectral reflection prediction model. The pattern is thus invisible under nonspecular viewing conditions but becomes visible under specular viewing conditions, thanks to the presence of the silver ink. The MetallicArt technique works well when the leafing effect is strong, i.e., when the metallic particles present in the metallic ink align themselves horizontally along the surface of the ink film. This effect is, however, reduced when printing with a web offset press. Due to the speed of such presses and to avoid smearing the press, the metallic ink must be dried as soon as printed, a process called UV curing. This fixates the metallic particles on the paper substrate before the leafing effect can fully take place, reducing the specular reflection of the metallic ink. Figure 1 shows the impact of UV curing. In Fig. 1(a), the leafing effect is reduced due to UV curing and the metallic particles cover much less of the surface than in Fig. 1(b) where the ink is dried without UV curing. This reduced specular reflection becomes a problem in regions largely covered by yellow. The yellow ink is not entirely transparent and slightly scatters light, showing a surprisingly strong specular reflection. This specular reflection competes with the specular reflection of the metallic ink and reduces the visibility of the embedded pattern under specular viewing conditions. In this contribution, we first propose a simplified CMYS spectral prediction model for the MetallicArt technique and compare its prediction accuracy with the accuracy of the more complex CMYS model described in Ref. 1. We, then, show that the specular reflection of the silver ink is not visible in the presence of yellow. We, therefore, propose an algorithm that minimizes the coverage of yellow in order to increase the visibility of the embedded metallic patterns under specular viewing conditions. EXPERIMENTAL DETAILS All the experiments are performed on a Müller Martini Concept web offset press using the cyan (43UE5200), magenta (42UE5200), and yellow (41UE5200) standard process inks from StehlinþHostag (NewV set MFX) and metallic silver ink SICPA 367701. The paper used is a mat coated Artic silk of 170 gr=m2. We use classical rotated halftone screens at 150 lpi imaged at 2400 dpi with angles of 75 , 15 , 0 , and 45 for the cyan, magenta, yellow, and silver inks, respectively. ~ IS&T Member. Received Apr. 18, 2011; accepted for publication Sep. 30, 2011; published online Dec. 23, 2011 1062-3701/2011/55(6)/060506/9/$20.00. J. Imaging Sci. Technol. Nov.-Dec. 2011 060506-1 Journal of Imaging Science and TechnologyV 55(6): 060506-1–060506-9, 2011. VC Society for Imaging Science and Technology 2011 Reprinted with permission of IS&T: sole copyright owners of the Journal of Imaging Science and Technology. We performed measurements of the specular reflectance factors using a (0 :0 ) geometry using the following Ocean Optics products. The light source is DH-2000-BAL’s halogen lamp. We used a 230 lm fiber optic reflection probe (QR2307-XSR) with a RPH-1 probe holder to illuminate the sample and collect the reflected light. The reflected light was measured with a Maya 2000 Pro spectrophotometer. Spectralon was used as the diffuse reference white. All reflectance factors were calculated in respect to this reference white. INK SPREADING ENHANCED YULE–NIELSEN MODIFIED SPECTRAL NEUGEBAUER MODEL (IS-YNSN) In order to preserve the original color of a CMY patch under nonspecular viewing conditions, while adding as much silver ink as possible, we use a spectral reflection prediction model. This model must be able to accurately predict the nonspecular reflectance of any CMYS patch. The original MetallicArt technique is based on an extension of the Clapper–Yule spectral reflection prediction model. This extension incorporates special parameters to take into account specificities of the silver ink such as large ink trapping and a raised metallic reflection spectrum in the presence of other inks. These specificities increase the complexity of the model. We further refer to this model as the Clapper–Yule based MetallicArt model (CY-MArt). We propose instead to use the simpler Yule–Nielsen model enhanced with our ink spreading model. We first present the Yule–Nielsen modified Spectral Neugebauer model (YNSN) before describing the ink spreading model. The prediction accuracy of the resulting IS-YNSN is then compared to the prediction accuracy of the original CY-MArt model. The Yule–Nielsen Modified Spectral Neugebauer Model (YNSN) One of the first color prediction models is the Neugebauer model. In its original form, it predicts the RGB values of a color halftone patch. Yule and Nielsen modified this model to account for optical dot gain, Yule and Colt applied it for CIE-XYZ tristimulus values, and Viggiano extended it to the spectral domain. This model, now known as the Yule–Nielsen modified Spectral Neugebauer model (YNSN), predicts the reflection spectra of color-constant patches, whose ink coverages are given. The equation used to perform a prediction is the following: