Comparability of 3D Scan Eztracted Measurements and Traditional Measurements –A practical application of the ISO 20685 Standard
In the recent decade 3D body scanning has become a standard technology for anthropometric surveys performed world wide and some expect that body scanning may replace manual surveys completely in the future. 3D body scanning provides multiple benefits compared to the traditional set of one-dimensional anthropometric measures, since the complete 3D-shape of the subject is recorded and available for subsequent statistical analysis. Nevertheless, past results and anthropometrical knowledge from traditional studies form the basis for observations on anthropometrical changes in populations over long periods of time. Thus, comparability of 3D scan extracted measurements to traditional measurements is a fundamental issue for bridging between traditional one-dimensional anthropometry and 3D surface anthropometry methodologies. ISO 20685 made an important step towards standardization. But, due to methodological differences in traditional measurement definition and 3D body scan extracted measurements and due to the previous lack of standardization methods, direct comparability of body measurements and scan extracted measurements on an individual basis is currently hard to achieve. Some of the methodological differences are i) many of the traditional measurements are based on palpated bony landmarks; this information is not available in the 3D surface scans, ii) many different postures are used in the traditional measurement process (e.g. ISO 7250 uses more than 10 different postures) while this is not practical for in scanner based surveys and iii) automatic measurement extraction routines can not unambiguously be defined on the basis of the standardized measurement definitions like in ISO 7250 or in ISO 8559 due to the lack of precise algorithmic descriptions. However, ISO 20685 provides an appropriate procedure to validate scan derived measurements against manual measurements by defining acceptable tolerances for the variation of the differences. This paper extends the notion of comparability by taking systematic deviations identified during the measurement validation process directly into account and using these to establish comparability and compatibility to existing traditional anthropometrical data. Once these relations are determined they can be used to characterize the conditions and significance of scan extracted measurements relative to traditional data for each of the measurements included in the scanner based survey taking e.g. the ISO 20685 limits into account. The method was applied in a measurement validation study performed with the VITUS/Smart scanning system in a U.S. Army Clothing Issue Facility within the USFIT project. The results observed are used to illustrate the proposed method and demonstrate its applicability to the measurement set defined for the study. These results suggest that comparability of scan and manual anthropometry can be established under defined conditions and restrictions. This method allows comparison of results within and between different scan based and manual surveys until more general standardization become accessible for implementation in the measurement extraction of future scanning systems.
Rainer Trieb Andreas Seidl Claire C. Gordon Bruce Bradtmiller Brian D. Corner Steven P. Paquette
Human Solutions U.S. Army Natick Soldier RD&E Center Anthrotech
国际会议
17th World Congress on Ergonomics(第十七届国际人类工效学大会)
北京
英文
1
2009-08-09（万方平台首次上网日期，不代表论文的发表时间）