Color Mixture Model Based on Spatial Frequency Response of Color Vision

Abstract

This paper proposes an analytical model of color mixtures. The model considers a color mixture as follows; color signals of a sample image attenuate and overlap with each other in human color vision. First, the characteristics of the color signals in a frequency domain are expressed with a Fourier coefficient. Secondly, the features of human color vision are formulated in a frequency domain by a Gaussian low pass filter. Thirdly, a relational model of human vision and looking at sample images with mixed colors is proposed. Additionally, the color mixture is qualitatively discussed by using the proposed model. As a result, when the visual distance is longer, the waveform of the color signal is broader in the time domain, and the color mixing area increases according to the visual distance. Finally, subjective evaluation experiments were conducted to verify the validity of the proposal. Both the analytical and experimental visual distances become longer as the pitch becomes larger. However, the maximum gap of 6 m is within the absolute value in these results. The proposal can explain qualitatively the process of this phenomenon by using the arrangements of color signals and the visual distance between the sample and observers as parameters.