Research on Product Image Styling Design Method Based on Neural Network and Genetic Algorithm

摘要

On the study of kansei engineering and product image evaluation, a product image styling design method based on neural network and genetic algorithm is presented. Firstly, according to the findings of kansei image, image evaluating pattern with neural network is constructed; and then, according to consumers' kansei image demands and cooperated with evaluating pattern of neural network as genetic filial generation's fitness function, the model of product image styling design with genetic algorithm is developed. It can automatically reason out the conceptual image styling product which will well satisfy consumers' psychical demands. Finally, the colour design of MP3 is taken to examine and certify the verification of the method. However, the proposed method is equally applicable to the styling design of other products.