The print quality of prints performed with flexographic printing technology is influenced by various parameters such as viscosity of the printing inks, printing substrates, plates, anilox rolls, etc. The aim of this work is to analyze the influence of ink viscosity, printing plate and printing base on selected properties of print quality, such as optical density of full tone area and the increase of tonal value (TVI). Additionally, the printed dots on 5, 15, 30, 50 and 75% of half tone area were investigated using digital microscopy. The least square fitting method in a matrix form was successfully used in order to confirm the influence of printing process parameters on print quality. The results obtained reveal the significant influence of the printing plate and printing substrate characteristics and the lesser influence of printing ink viscosity for 40–100% coverage area. The values of the optical densities of full tone areas are mainly influenced by the properties of the printing plate and the printing base. Furthermore, the optical density decreases when the value of flow time decreases, which is related to the lower thickness of dried ink film. The TVI on light tones (0 to 20%) is mainly influenced by the ink viscosity.

A machine learning-based application for estimating a reflected electromagnetic spectrum is presented in this paper. Since the results of the spectrum estimation by our previously implemented Cubic Hermite interpolation method had some drawbacks, and in line with the increasing application of machine learning in sensor technology, the new hypothesis is put forward, based on the assumption that better predictions could be achieved by introducing ANN (Artificial Neural Networks). The input data of such networks consist of six measured values which represent the intensities of reflected light at specific wavelengths, while the output is formed out of thirty-six points which are used to predict the shape of the spectral curve in the range of 380-730 nm (to obtain a resolution of 10 nm). Colorimetric capabilities of the proposed method are compared with a commercial spectrophotometer, as well as with our previously obtained results using the Cubic Hermite interpolation. Machine learning-based method has proven to be an intelligent solution for the spectrum estimation, and the drawbacks from the previously used method were reduced significantly.