Research on Product Image Form Design Based on ANFIS

Product image form design, which focuses on customers´ psychological demands, is arousing attention increasingly. This paper presents a novel approach of customer-oriented design for translating customers´ kansei image into product design elements. The most influential form elements are identified through rough set theory. Based on this, the relationship between the key form elements and customers´ kansei is constructed by adaptive neuro-fuzzy inference system (ANFIS) using MATLAB. To validate the prediction performance of ANFIS, fuzzy logic and back propagation neural network models are utilized. An experimental study of glasses design is conducted based on the proposed method, and the results suggest that ANFIS has a good prediction performance.