Impact of a transparent layer added on a semi-tone print

The addition of a transparent coating layer over a halftone print can significantly change its color, even if the layer is perfectly clear and of similar roughness as the uncoated print. The objective of the present work was to understand, model, and evaluate the impact of the optical phenomenon at the origin of this color change for color management purposes in the printing industry.This color change is caused by the interreflections of light in the coating layer between the diffusing substrate and the coating-air interface. This lateral light propagation has a very specific ring-shaped point spread function caused by the angular dependency of the Fresnel reflectance at the interface between the coating layer and air. This ring-shaped lateral propagation enables light to meet various inked areas of the printed halftone pattern and be absorbed by them, which induces a darkening and a saturation of the color which can be seen as an additional optical dot gain for coated prints.

We proposed a reflectance model for this optical phenomenon (that we have called halation phenomenon) which relies on multiple convolutions between the ring-shaped point spread function, function of the thickness of the coating layer, and the spatial and spectral intrinsic reflectance of the non-coated print.

Two macroscale experiments were performed for verification of the model: the reflectance factor of centimetric and millimetric printed halftone patterns coated with various coating thicknesses were measured. These experiments enabled to validate the theory behind the halation phenomenon, both in the spectral and in the spatial dimensions.Simulations with the multi-convolutive model allowed to evaluate the impact of various printing parameters in the color change caused by the addition of a coating layer. In particular, simulations through a microfacet-based BRDF showed that the roughness of the coating-air interface had almost no impact on the halation phenomenon. The multi-convolutive model may then be applied to both glossy and matte coatings by adjusting, for each surface roughness, a parameter in the model which accounts for the light portion externally reflected by the interface towards the sensor. The halation phenomenon has been studied at the microscale for electrophotographic and inkjet prints. At this scale, the intrinsic reflectances of prints are impacted by mechanical and optical dot gains.

To calibrate the multi-convolutive model, an apparatus was developed to measure the spatial and spectral reflectance factor of the prints: a multispectral microscope. This microscope captured pictures through rather small wavelength bandpass optical filters (25 nm), allowing to characterize the prints both spectrally and spatially. This apparatus was calibrated to find an agreement between microscale measurements, performed with the multispectral microscope, and macroscale ones, performed with a spectrophotometer. This setup allowed to calibrate the multi-convolutive model and to evaluate its accuracy for microscale prints coated with a 25 µm-thick lamination layer. The color difference between the predictions given by the model and the reflectance factor measurements of coated prints was in average ΔE=0.95±0.39 for electrophotographic prints and ΔE=0.68±0.28 for inkjet prints. These fairly accurate results make the multi-convolutive model a rather promising tool for color management in the printing industry. As the multispectral microscope used to calibrate the multi-convolutive model is a rather rare instrument, preliminary simulations are presented which attempt to use classical RGB microscopic measurements to calibrate the model.

Knowing how the halation phenomenon impacts the color of various halftone patterns also paves the way towards security applications. An application is presented where it is shown that images can be hidden and revealed only by adding or removing a clear coating layer.

Thesis available on: https://cnrs.hal.science/LGP2/tel-04414909v1