Page 45 - Anthropometry, Apparel Sizing and Design

Anthropometric methods for apparel design: Body measurement devices and techniques 41

2. Calculate the error for all subjects. The error is defined as the difference between the scan-
derived measurement and the measurement by the skilled anthropometrist (scan-derived
measurement minus manual measurement).
3. Calculate the 95% confidence interval of the mean error. The lower and upper limits of the
95% confidence interval are calculated as the mean error 1.96 the standard error. The
standard error is calculated as the standard deviation divided by the square root of N.
4. When the following equations are satisfied, the two types of measurements are considered
sufficiently similar: the maximum allowable error<lower limit of the 95% confidence
interval of the mean error, and the upper limit of the 95% confidence interval of the mean
error<+the maximum allowable error. The maximum allowable error is 4mm for height,
small girth, and body breadth measurements; 5mm for segment length and body depth mea-
surements; 9mm for large girth measurements; 2mm for foot measurements and head mea-
surements including the hair; and 1mm for hand measurements and head measurements not
including the hair.
The procedure is explained using an example of back neck height (cervical height).
Back neck height was measured for 74 subjects (39 females and 35 males) by a
bodyline scanner (Hamamatsu Photonics K.K., Hamamatsu, Japan) and by a skilled
anthropometrist. The error was calculated as the difference between the scan-derived
measurement and the traditional measurement for the 74 subjects. Means and
standard deviations are shown in Table 2.2. The lower limit of the 95% confidence
interval is calculated as mean error 1.96S.E.¼mean error 1.96S.D./√N ¼
5.6–1.96 4.9/√74¼4.48 [mm]. The upper limit of the 95% confidence interval is
calculated as mean error+1.96S.E.¼5.6+1.96 4.9/√74¼6.72 [mm]. The maxi-
mum allowable error of height measurements is 4mm. As the upper limit of the
95% confidence interval is larger than the maximum allowable error, the results do
not satisfy the condition, and we conclude that the scan-derived and traditional measure-
ments are not comparable in this example; this bodyline scanner systematically gives
larger values. Fig. 2.9 shows the relationship between the 95% confidence interval
of the mean error and the maximum allowable error in this example.

2.3.5 Accuracy of scan-derived 3-D measurements
ISO 20685-2:2015 establishes a protocol for evaluating the performance of 3-D body
scanners. The basic idea is to measure an object with known shape and size and com-
pare the measured results with the actual values. An artifact (ball) calibrated using a
coordinate measuring machine (CMM) that is traceable to the international standard

Table 2.2 Comparison of scan-derived and traditional measurements
of back neck height (cervical height) (N ¼74) (unit: mm)
Scan-derived Traditional
measurement measurement Error
Mean 1388.2 1382.6 5.6
S.D. 76.9 76.1 4.9