摘要
在采用三维人体扫描技术精确、快速地采集到人体全身数据后,对所获得的数据进行探索性分析、配对样本t检验等预处理,再用因子分析和层次聚类分析相结合的数据挖掘方式提取颈椎点高、总肩宽、胸围这3个识别人体体型特征的变量,并用其进行K-means人体体型聚类分析。研究结果表明,所提出的方法既可以作为实体商店或网购服装的参考标准,也可以用于了解某地区的人体体型特征,进而改良服装结构设计。
Whole body data of young females were acquired by three-dimensional body scanning technology. After data preprocessing including exploratory analysis and paired samples t test, a kind of data mining technique combing factor analysis with hierarchical clustering was employed to extract the key body measurements for identifying the human body types. Afterwards, the human body types were analyzed by K-means clustering based on the features extracted. The results showed that the method proposed in this paper could not only be used for purchasing garments in daily life, but also for improving the garment patterns based on understanding the human body characteristics in certain areas.
引文
[1] 湛群,赵楠翔.基于改进蚁群算法的非接触人体测量中体型聚类分析[J].赤峰学院学报(自然科学版),2014,30(15):56-57.
[2] 许平山,高云兵.安徽省小学生体型分类及号型设定研究[J].赤峰学院学报(自然科学版),2014,30(15):61-63.
[3] DEEPTI G,NORSAADAH Z.Anthropometry,apparel sizing and design[M].London:Woodhead Publishing,2014:95-119.
[4] 刘爱玉,田志鹏.SPSS数据分析教程[M].北京:北京大学出版社,2017:196-199.
[5] 卢纹岱.SPSS for Windows 统计分析[M].3版.北京:电子工业出版社,2008:140-143.