RESEARCH ARTICLE


Alignment Influence on 3D Anthropometric Data Clustering



Jianwei Niu1, Zhizhong Li 1, *, Gavriel Salvendy 1, 2
Department of Industrial Engineering, Tsinghua University, Beijing, 100084, China.


Article Metrics

CrossRef Citations:
1
Total Statistics:

Full-Text HTML Views: 744
Abstract HTML Views: 1233
PDF Downloads: 201
Total Views/Downloads: 2178
Unique Statistics:

Full-Text HTML Views: 338
Abstract HTML Views: 714
PDF Downloads: 145
Total Views/Downloads: 1197



Niu et al.; Licensee Bentham Open

open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

* Address correspondence to this author at the Department of Industrial Engineering, Tsinghua University, Beijing, 100084, China; Tel: +86-10- 62773923; Fax: +86-10-62794399; E-mail: zzli@tsinghua.edu.cn


Abstract

Shape analysis and comparison is important to sizing system for the design of many products which require close fitting contact with the human body. However, the choice of the alignment reference can profoundly influence the shape analytical results. The objective of this case study is to demonstrate that the statistical results of the threedimensional (3D) upper heads could be different if different alignment approaches are used for the data analysis. Taking a data set of 432 upper heads as an example, this paper addressed the influence on data analysis of two alignment approaches, i.e., aligned at the centroid and aligned at the head top. K-means clustering was applied on block-based distance vectors of the upper head samples to classify the population into categories based on their shapes. Cluster membership variation of different alignment methods was examined. Results indicated that the reference can greatly influence k-means clustering results of 3D anthropometric data. Thus, alignment reference should be carefully chosen according to the specific requirements of an application.

Keywords: Three dimensional anthropometry, clustering, sizing, alignment.