Thesis (Ph.D.)--Chulalongkorn University, 2013

The current colour system for printing specifies colours using percentages of dot areas of each primary colour, i.e. cyan, magenta, yellow and black. However, this system does not conform with the colour measurement system that relates to additive colour-mixing with red, green and blue as primaries. Thus, a colour system based on colour perception of prints will provide better correlation between the colour measurement system and the printing system. This study proposed colour strength as the colour perception parameter used to connect these two systems. The aim of this study was to devise a colour system based on colour perception for primary colours in printing. The relationships between the dot area percentages of primary colours and colour strength were obtained through a series of experiments. The colour samples were printed with the offset printing system. The colour strength values of the grey samples in the reference scale were quantified by the magnitude estimation method. For the primary colour samples (C, M, and Y), the colour strength values were obtained from matching the samples to the reference scale. The second-order polynomial equations for computing the colour strength value from the dot area percentage of C, M, Y, and K and vice versa were derived. The colour strength values obtained from these equations and CIEL*a*b* values were used to derive the colour strength model. The colour system proposed in this study used CIEL*a*b* values as the input data to predict the colour strength values, then from the colour strength values to predict the dot area percentages of primary colours. The performance of the system was tested with secondary colour samples (RGB). It was found that the percentages of colour strength prediction errors for red, green and blue were 15.92, 14.07 and 43.10, respectively. The percentages of %dot area prediction errors for red, green and blue were 6.28 (average errors of Y and M), 3.90 (Y and C) and 3.28 (C and M), respectively. Since the agreements between observers’ visual data of colour strength were approximately 20% errors, the proposed colour system was considered to give good predictions.