Research on image design of Fujian paper-cut pattern based on Kansei engineering and WOA-BP neural network

Abstract

In order to design Fujian paper-cut patterns that meet the perceptual needs of consumers and better inherit and develop them in modern society, a Fujian paper-cut pattern image design method based on perceptual engineering and the Whale Optimization Algorithm optimized BP neural network (WOA-BP) neural network is proposed. First, based on the theory of Kansei engineering, six representative paper-cut pattern samples and their main modeling features were determined through questionnaire survey, multi-dimensional scaling analysis, cluster analysis, and analytic hierarchy process. Second, the semantic difference method is used to obtain the perceptual image evaluation value of the representative paper-cut pattern, and combined with the principal component analysis method, the representative image vocabulary is extracted. Finally, the WOA-BP neural network is used to construct the mapping relationship between consumers' perceptual images and paper-cut pattern modeling features, and calculate the paper-cut pattern modeling code combinations corresponding to consumers' image needs. At the same time, the paper-cut pattern design using the image vocabulary ‘modern-traditional’ as an example verifies the feasibility of the method in this article. Compared with the existing method of designing paper-cut patterns based solely on the subjective experience of the designer, the method of this article can correlate the perceptual needs of consumers with the corresponding modeling characteristics of paper-cut patterns, making the design of paper-cut patterns targeted, precise and intelligent.