Modeling Design Method and Application of Cheongsam Based on BP Neural Network
In order to solve the problem that the design process of cheongsam is complex and the consumer demand is difficult to meet,this paper deconstructs and codes the cheongsam shape based on Kanesei engineering and modular design concept,deconstructing it from the collar type,skirt type,sleeve type,slit and pendulum type,and screening 10 pairs of perceptual evaluation words.The questionnaire was used to quantify the user's perceptual image,factor analysis and BP neural network were used to explore the quantitative relationship between the user's perceptual intention and the chi-pao shape,and the corresponding design code was simulated according to the prediction target and applied to the targeted cheongsam shape design.The results show that this method can effectively improve the design efficiency of cheongsam modeling,realize the integration and application of user's perceptual image in cheongsam design,and reduce the sample size of measuring user's perceptual image.This method is used to complete the design of the noble style cheongsam,and the feasibility of the method is verified.
NETL
NSTL
万方数据
