Exploring Color Features with Nasnet CNN and Improved Water Cycle Algorithm for Product Emotional Design

This research investigates the multidimensional domain of color image optimization design for emotional product color design, in this case, tricolor product color schemes. By introducing novel color-area ratio modes and with the use of advanced technologies, including the combination of a Nasnet convolutional neural network and an optimized Water cycle algorithm, this research attempts to enhance color design towards a higher emotional effect. Through a systemic process of color scheme generation and evaluation, five perceptual characteristics are identified, leading to the development of emotional evaluation models. Integration of different methodologies allows multiobjective optimization to be attained, leading to better color harmony and users’ emotional preferences. The results of the validation demonstrate the effectiveness of the proposed optimization method. The effectiveness of the study is proved by applying it to children’s electric car color palettes so that its success in real application is evident. The study helps in product emotional design by providing an effective and efficient approach to color optimization.