基于BP神经网络的敦煌图案在旗袍中的感性应用  

Perceptual application of Dunhuang patterns in cheongsam based on BP neural network

作　　者：方思涵 薛媛[1] FANG Sihan;XUE Yuan(College of Clothing and Art Design,Xi′an Polytechnic University,Xi′an,Shaanxi 710048,China)

机构地区：[1]西安工程大学服装与艺术设计学院,陕西西安710048

出　　处：《毛纺科技》2025年第3期86-93,共8页Wool Textile Journal

基　　金：教育部人文社科基金项目(21YJA760079)。

摘　　要：为提高敦煌图案在旗袍中的设计效率,满足消费者对旗袍图案的感性需求,提出基于BP神经网络的敦煌图案在旗袍中的感性设计方法。筛选7类符合旗袍风格的敦煌图案,绘制16款旗袍刺激图,利用语义差异法设计调查问卷,获取消费者在12个维度对刺激图的感性评价值,利用因子分析法对评价值进行数据分析,选出3个感性因子,从3个维度分析敦煌图案的设计要素并对其进行模块编码,构建消费者对旗袍的感性评价与敦煌图案的设计要素之间的关联模型,训练并验证该模型的准确性。结果表明:敦煌图案的类别、单元图案尺寸、排列方式的不同组合会使消费者展现出不同的感性心理;该BP神经网络模型具有可行性,可指导敦煌图案在旗袍中的感性设计。In order to improve the design efficiency of Dunhuang patterns in cheongsam and meet consumers′ perceptual demand for cheongsam patterns,a perceptual design method of Dunhuang patterns in cheongsam based on BP neural network was proposed.Seven types of Dunhuang patterns conforming to the style of cheongsam were selected,16 types of cheongsam stimulus maps were drawn,and a questionnaire was designed using the semantic difference method to obtain consumers′ perceptual evaluation value of the stimulus maps in 12 dimensions.Factor analysis was used to analyze the data of the evaluation value,and 3 perceptual factors were selected to analyze the design elements of Dunhuang patterns from 3 dimensions and then module coding was performed.The correlation model between consumers′ perceptual evaluation of cheongsam and the design elements of Dunhuang patterns was constructed,and the accuracy of the model was trained and verified.The results show that the different combinations of Dunhuang pattern category,unit pattern size and arrangement will make consumers show different perceptual psychology.The BP neural network model is feasible and can guide the perceptual design of Dunhuang patterns in cheongsam.

关 键 词：感性工学 旗袍 敦煌图案 因子分析 BP神经网络 

分 类 号：TS941[轻工技术与工程—服装设计与工程]