To improve the consistency between color semantic expression and user emotional needs in graphic design, a graphic color matching method based on emotion-driven and digital visualization color mapping is proposed. This method quantifies users’ ratings of pleasure and activation through a self-assessment model to form a two-dimensional emotional space. The Formula: see text-means algorithm extracts color features from color palettes. A multi-task lasso regression model is used to jointly train an emotion prediction model. The L1/L2 mixed regularization is introduced to screen key features. In the optimization phase, multi-objective functions are used to synchronously optimize emotional bias and color distinguishability. A color mapping algorithm based on color gamut boundary intersection points is designed to reduce the cross-device color offset rate. Experiments showed that the model had high reliability in predicting pleasure and activation, with Cronbach’s alpha reaching 0.9037 and 0.9045, respectively, and mean square error converging to 0.012. User testing showed that the optimized color palette’s emotional coordinates reduced the deviation from the target emotion by 42%. In scenes such as film and television stylization and data visualization, the efficiency of color style conversion was increased by 35%. The color palette achieved an emotional matching accuracy of 89% while maintaining color harmony. The study addresses the limitations of traditional discrete emotion mapping, providing theoretical support for interdisciplinary research in color psychology and computer graphics.
Yiran Kong (Tue,) studied this question.