Sensory Styling Design of Physiotherapy Beds based on BP Neural Network

Abstract

There is a significant gap between the form of current physiotherapy beds and users' perceptual images. By employing techniques such as computer graphic design and logical operations, analyzing the mathematical relationship between user needs and design elements contributes to enhancing the scientific nature of product design and making the design process more rigorous. This paper, based on an analysis of the product image and design process, employs a comprehensive fuzzy evaluation method to identify representative perceptual vocabulary for physiotherapy beds. The KJ method and Delphi method are utilized to select necessary samples. A morphological analysis matrix is established using the morphological analysis method, and a comprehensive decision-making model for product design is constructed using a BP neural network on the MATLAB platform. Through training the BP neural network on physiotherapy bed products, it becomes possible to predict the perceptual evaluation of product design, achieving a quantification of the design. This provides valuable support for the appearance design of physiotherapy bed products and significantly enhances the efficiency of designers' work.