紡織工程畢業(yè)外文翻譯--使用三維身體掃描，對織物懸垂性評估的調(diào)查

1、<p><b>　　譯 文</b></p><p>　　原文題目: An investigation of the assessment of fabric drape using three-dimensional body scanning </p><p>　　譯文題目: 使用三維身體掃描，對織物懸垂性評估

2、的調(diào)查 </p><p>　　學(xué) 院: 紡織與材料學(xué)院 </p><p>　　專業(yè)班級: 紡織工程 </p><p>　　學(xué)生姓名: &l

3、t;/p><p>　　學(xué) 號: </p><p>　　An investigation of the assessment of fabric drape using three-dimensional body scanning</p><p>　　Tannie Mah and Guowen

4、 Song*</p><p>　　Department of Human Ecology, University of Alberta, Edmonton, Canada</p><p>　　(Received l/April 2008;final version received 13 August 2008)</p><p>　　This investigati

5、on explores how information on air gaps obtained from three-dimensional (3D) body scanning can be used to evaluate the drape of fabrics. Three-dimensional images of draped fabrics differing in physical and mechanical pro

6、perties were acquired through 3D body scanning and the air gap distances and distribution between the outside edge of a cylinder and the fabric determined. Results demonstrate an inverse relationship between the overall

7、average air gap distance and the amount of f</p><p>　　Keywords: fabric drape; 3D body scanning; air gap; drape coefficient</p><p>　　Introduction</p><p>　　Drape is one of the propert

8、ies that can influence the aesthetic perception of a textile product. Not only does drape affect consumer acceptance of a textile product, but it also plays an influential role in the design, production, and retailing st

9、ages of that item. Because drape determines the adjustment of clothing to the human silhouette (Kenkare & May-Plumlee, 2005), garment fit, mobility, and human comfort are also affected (Hunter & Fan, 2004). Drape

10、 is therefore an important factor, determ</p><p>　　The drape behavior of fabrics is complex. Each fabric has a distinct set of properties that causes it to drape in a characteristic way (Kenkare & May-Pl

11、umlee, 2005). Not only do drape shapes differ across fabrics, but a single fabric alone may take on different drape configurations with</p><p>　　repeated draping as well (Jeong, 1998). Variations in drape ar

12、e due to the complicated interactions between optical, physical, and mechanical fabric properties, as well as other factors.</p><p>　　Despite its complexity, methods that evaluate the drape of fabrics have b

13、een developed. Research in the area has generally followed one of the following three approaches: (1) the objective measurement of a fabric's material properties, namely shear, bending, and weight, and correlation of

14、 these values with drape values; (2) the objective measurement of drape attributes using the drapemeter, specifically the drape coefficient and node properties; and (3) the subjective evaluation of fabric drape </p>

15、;<p>　　development of fabric drape prediction models and simulation techniques (Stylios & Wan, 1999). More recently,May-Plumlee et al. (2005) used a three-dimensional (3D) body scanner to investigate the virtu

16、al draping of garments,taking variations in fabric mechanical properties into account.</p><p>　　The subjective evaluation of drape involves either the kinesthetic or visual judgment of human raters. Evaluato

17、rs use polar words on value scales, rank the amount of and/or preferred drape on ordinal scales, or make comparisons of specimens against a reference fabric. Tedious mea-</p><p>　　surement procedures and lac

18、k of knowledge in conducting objective tests and interpreting results make the subjective assessment of drape common in textile and apparel industries (Gider, 2004). However, the strong subjective quality of drape sugges

19、ts that accurate subjective assessment of this property is difficult. Results may be biased and inconsistent due to the vagaries of personal preference, human perception, and fashion change. Subjective methods have there

20、fore generally been deemed unreli</p><p>　　Research since the 1930s resulted in the development of objective techniques and quantitative parameters for the assessment of fabric drape. Recognition that fabric

21、 mechanical properties such as stiffness had a significant effect on drape prompted researchers to use instruments that</p><p>　　measured these properties as a means of indirectly evaluating drape. In the ob

22、jective two-dimensional (2D) cantilever method (Peirce, 1930), a rectangular strip of fabric is placed on a horizontal platform and extended over the edge until the overhanging end contacts an angled platform. The longer

23、 the projected bending length, the stiffer the fabric,</p><p>　　and the lower the drape, and vice versa. The American Society for Testing and Materials (ASTM) D 1388 Standard Test Method for Stiffness of Fab

24、rics uses this procedure and includes alternative 2D techniques such as hanging loop tests (ASTM, 2007).</p><p>　　Two-dimensional tests are relatively easy to perform, have good reproducibility, and are comm

25、only used in industry or for experimental and theoretical drape studies.However, 2D instruments cannot discriminate between papers and fabrics that have the same stiffness value (Kenkare & May-Plumlee, 2005). Even if

26、 these sheet materials possess the same bending lengths, it is unlikely that paper will drape in the same manner as fabric. Also, bending resistance is measured in only one direction at a tim</p><p>　　anisot

27、ropic behavior of fabrics is not captured by 2D tests. Fabric drape is a 3D phenomenon and cannot accurately be described by 2D measures.</p><p>　　Fabric drape measurement took a significant step forward wit

28、h the development of the drapemeter by Chu, Cummings, and Teixeira (1950) and was later improved by Cusick (1965, 1968). A drapemeter, an instrument capable of distorting a fabric in all three dimensions at once, consist

29、s</p><p>　　of two horizontal circular disks in between which a circular fabric specimen is sandwiched. The portion of the fabric not held between the plates drapes freely over the edge of the lower disk, fro

30、m which a light source below causes a shadow of the draped fabric to be cast onto a paper above the apparatus. The outline of the shadow is traced on the paper and the drape coefficient determined.</p><p>　　

31、According to the British Standard (BS) 5058 Method for the Assessment of Drape of Fabrics (1973), the drape coefficient is the "percentage of the total area of an annular ring of fabric obtained by vertically projec

32、ting the shadow of the draped specimen" (p. 4/29). The drape coefficient, an objective and direct measure of fabric drape, can be</p><p>　　obtained through various means including area and mass measurem

33、ents. A drape coefficient between 25% and 50% indicates a drapable fabric and a drape coefficient greater than 75% a less drapable fabric (Collier & Epps, 1999).</p><p>　　Although the drape coefficient i

34、s a stable and reproducible parameter for characterizing fabric drape (Mizutani, Amano, & Sakaguchi, 2005) and is the most widely used estimating value for quantifying drape in textile and apparel industries (Kenkare

35、 & May-Plumlee, 2005), this number alone does not provide a complete description of fabric drape (Heafie, 1969). The drape coefficient only provides a measure of the degree of drape, without any information about the

36、 appearance of a draped fabric. Know</p><p>　　the form of a draped fabric provides a more complete and useful description, especially as draped fabrics are usually viewed from the side in real life. Furtherm

37、ore, because it is possible for two fabrics with similar drape coefficients to have different draped appearances (Kenkare & MayPlumlee, 2005; Stylios & Zhu, 1997), additional parameters that could provide more in

38、formation about the appearance of a draped projection were needed (Mizutani, Amano, & Sakaguchi, 2005).</p><p>　　Because drape appearance can be better envisioned from the number of fabric folds, the num

39、ber of nodes is another commonly used parameter (Jeong, 1998). According to Chu, Platt, and Hamburger (1960), a drape diagram (the projected 2D simplification ofa 3D draped fabric) has three important features: the area,

40、 number of nodes, and shape of the nodes. When used together, a more complete description of fabric drape is possible. Whereas very stiff fabrics sag only slightly without forming any defini</p><p>　　Other p

41、arameters have also been used to supplement the drape coefficient. The drape~tistance ratio was proposed by Jeong (1998) as an alternative to the drape coefficient. Stylios and Powell (2003) used the drape coefficient an

42、d the number, depth, and evenness of folds to characterize drape. In addition to the drape coefficient, Stylios and Wan (1999) used the fold number, fold variation, and fold depth index to</p><p>　　better re

43、present how humans interpret drape aesthetically. The average of the angles from the point of overhang to the free end of a fabric at the maximum and the minimum node locations was proposed by Stylios and Zhu (1997) to b

44、e a better representation of drape than the drape coefficient. Also, Mizutani et al. (2005) defined a new parameter to express the simplicity or evenness of a draped projection.</p><p>　　Variations of the or

45、iginal drapemeter and of obtaining the drape coefficient and other drape parameters have been developed, including the photovoltaic cell method (Collier, 1991), the drape elevator technique (Mizutaniet al., 2005), and th

46、e static-and-dynamic drape measurement system (Stylios & Zhu, 1997). With recent innovations in technology, image analysis systems have also been used (e.g. Jeong, 1998; Lo, Hu, & Li, 2002).1</p><p>

47、　　Fabric drape measurement techniques have advanced over the years and are continually evolving as technologies improve. However, although it has been realized that fabric drape is a 3D phenomenon, measurement technologi

48、es other than the conventional drapemeter have been limited</p><p>　　(Stylios & Zhu, 1997). Two-dimensional stiffness and 3D drapemeter approaches cannot fully capture the truly 3D form of a draped fabri

49、c. In this research, a 3D body scanner is used to characterize the drape of a series of fabrics. By visualizing and quantifying the air gap distances and distribution between a fabric and a cylinder through the examinati

50、on of cross-sections, the overall average air gap distance has been found to be a useful parameter for characterizing the drape of fabrics, and ca</p><p>　　traditionally used parameters. </p><p>

51、;　　Table 1. Fabric structure and physical properties</p><p>　　Experimental</p><p>　　Test specimens</p><p>　　Fabrics thought to represent a wide range of drape values were chosen and

52、 were not selected on the basis of their specific physical and mechanical properties. Denim cotton, finished acetate,2 flame-resistant (FR) cotton, linen, Nomex~,polyester, and silk fabrics were used. Table 1 outlines th

53、e characteristics of the test specimens.</p><p>　　For the 3D body scanning, test specimens were cut into 100-cm-diameter circles. This size allowed enough fabric overhang around the cylinder for a number of

54、cross sections to be taken along the length of the overhang. This larger size was also thought to better represent a more realistic clothing situation compared to the smaller specimen size used in the drapemeter approach

55、. One specimen was prepared for each fabric with three replications performed on each specimen.</p><p>　　To compare the drape values obtained from the new 3D body scanning approach with the traditional drape

56、meter method, the drape coefficient for each of the fabrics was calculated according to the BS 5058 test method for the assessment of fabric drape. Two 36-cm diameter circles were prepared for each fabric as specified in

57、 the standard, and three replications for each specimen performed.</p><p>　　Also, to see how the 3D body scanning results compared to the 2D stiffness results, ASTM D1388 for measuring the stiffness properti

58、es of fabrics using Option A—Cantilever Test was performed. Four specimens 2.5 cm x 20 cm were prepared for each of the warp and weft directions for each fabric. Each specimen was tested with the face side up and on both

59、 ends and with the face side down and on both ends for a total of four measurements per specimen.</p><p>　　Test apparatus</p><p>　　A 3D body scanner is a noncontact optical measurement system c

60、apable of rapidly generating a 360~ representation of the surface geometry of an object. Laser scanners project a line of horizontal light onto an object, which is reflected back into cameras that move vertically along t

61、he length of the scanning volume. Software uses the displacement of the light pattern to calculate the distance from the object to the camera, from which the data is inverted to produce a 3D representation (Kaufmann, 1&l

62、t;/p><p>　　Three-dimensional body scanning was conducted using a Vitus 3D whole body scanner (Human Solutions, Kaiserslautern, Germany). The scanner is connected to a computer equipped with the ScanWorX and Ant

63、hroscan 3D measurement software (Human Solutions). ScanWorX is used for the visualization, processing, and evaluation of scan information, whereas Anthroscan is used for the semiautomatic acquisition of measurements from

64、 the scanned object.</p><p>　　Fabrics were draped over a cylinder 120 cm tall and 22 cm in diameter within the scanning volume (Figure 1). This diameter was thought to better represent the size of a human

65、torso than those of a smaller diameter and</p><p>　　would also be useful for future research where the drape of garments over a human form would be determined. A pin affixed to the center of the top of the c

66、ylinder allowed the central placement of test specimens. An 18-cm diameter drapemeter designed as outlined in the BS 5058 method for assessing fabric drape was constructed for the measurement of the drape coefficient. A

67、cantilever bending tester as specified in ASTM D1388 was used to evaluate the stiffness of fabrics based on bending lengths. </p><p>　　Printouts of each cross-section were taken, and a 360~ protractor was ov

68、erlaid centrally on each image as illustrated in Figure 3. The air gap distance between the outside edge of the cylinder and the outside edge of the fabric was measured in 5~ increments around the circumference of the pr

69、otractor for a total of 72 measurements for each crosssection. The average air gap distance for each cross-section was then determined.</p><p>　　The procedure for determining the drape coefficient as outline

70、d in BS 5058 was followed with the exception that the draped image from the top view was captured with the 3D body scanner and printed instead of tracing the draped shadow onto paper by hand. The cut-and-weigh mass techn

71、ique as outlined in the standard was then performed. ASTM D1388 Option A--Cantilever Test for the measurement of bending length was followed.</p><p>　　Results and discussion</p><p>　　Examples of

72、 the 3D scan images of the draped fabrics are presented in Figure 4 from a single side view and the top view. The fabrics show different degrees of drape, draped appearances, and numbers and sizes of nodes. The drapabili

73、ty of fabrics increases from left to right, with fabrics on the left having larger average air gap distances than the fabrics on the right. The top views also demonstrate that as the drape increases, the overall average

74、air gap distances between the cylinder and the ou</p><p>　　Other parameters commonly used to describe the drape of fabrics such as the number and size of nodes can also be accounted for through 3D body scann

75、ing. The number and size of nodes affect both the average air gap distance and the appearance of a draped profile. In general, less drapable fabrics have larger air gap distances and fewer, larger nodes than more drapabl

76、e fabrics, which have smaller air gap distances and more nodes of smaller size. Figure 4 shows that the number of nodes increases fr</p><p>　　Table 2 shows the results of the air gap measurements for the dra

77、ped fabrics. Each cell represents the average of the 72 air gap distances taken across the cross-sectional elevations for each of the three replicates, with the averages, standard deviations, and combined total air gaps

78、calculated. The standard deviations are low, demonstrating the consistency and reproducibility of this approach. Figure 6 presents the combined average air gap distances across the cross-sectional elevations for eac</

79、p><p>　　A more detailed examination of the air gap distributions for each fabric is provided in Figure 7. Each bar represents the average of the 72 air gap distances taken at each of the 5, 10, 15, 20, 25, 30,

80、and 35 cm cross-sectional elevations for each of the fabrics. The finished acetate has the least drape across the cross sections according to the largest average air gap distances, and the silk has the most drape due to

81、the smallest average air gap distances across the cross-sections.</p><p>　　Figure 1. Schematic of draped fabric over cylinder within 3D body scanner.</p><p>　　Figure 2. Three-dimensional scan

82、images of air gap visualization procedure.</p><p>　　Figure 3. Image of merged cylinder and draped fabric crosssection with 360~ protractor.</p><p>　　Figure 4. Side and top view 3D scan images o

83、f draped fabrics.</p><p>　　Figure 5. Average number of nodes for draped fabrics.</p><p>　　Table 2 Average air gap distances (cm) acress cross-secitiaons for draped fabries</p><p>　

84、　It is clear from Figure 7 that as the cross-sectional elevations move from 5 cm to 35 cm, the average air gap distance from the outside edge of the cylinder to the outside edge of the fabric increases. This is expected,

85、 as the overhanging portion of the draped fabric widens toward the bottom of the fabric (Figure 4).</p><p>　　By analyzing a draped fabric's air gap distribution, such as in Figure 7, information about th

86、e appearance of a draped fabric can be gathered. To illustrate, Figure 8 shows three fabrics with different draped appearances. By analyzing the air gaps along the length of these fabrics, information about the draped fo

87、rm can be revealed. While draped fabrics with straight edges (Figure 8a) may demonstrate average air gap distances with a linear relationship across cross sections, fabrics with an outwa</p><p>　　Figure 9 he

88、lps to further illustrate this point. The finished acetate and linen fabrics demonstrate a nonlinear relationship between average air gap distances and crosssectional elevations along the length of the draped fabric.The

89、finished acetate has an outward curved line, suggesting a draped form with a convex appearance, and the linen has an inward curve, suggesting a draped profile with a slight concave appearance. The line between the end po

90、ints emphasizes the curvature of the data.</p><p>　　Figure 8 also demonstrates the inability of the drape coefficient to provide a complete description of draping behavior (Hearle, 1969). Although the three

91、fabrics would produce the same drape coefficient, the draped appearance of the fabrics might actually be quite different. The drape coefficient alone cannot provide these additional details about the draped form. The air

92、 gap distribution obtained from 3D body scanning can, however, differentiate between fabrics with the same drape coefficient </p><p>　　Figure 10 presents the average drape coefficient results determined acco

93、rding to the BS 5058 method for the assessment of fabric drape. A higher drape coefficient indicates a fabric with less drape, and vice versa. The finished acetate has the least drape according to its highest drape coeff

94、icient (84.0%), and the silk has the most drape based on its lowest drape coefficient (39.3~~. The drape coefficients of the denim cotton and Nomex--(73.3% and 71.7%, respectively) are quite close, indicating</p>

95、<p>　　Drape determination using the 3D body scanner may be better able to differentiate between fabrics of similar drapes than the drapemeter approach due to the larger size of the test specimens used. In the 3D bod

96、y scanning, the greater amount of fabric overhang around the cylinder relative to the drapemeter test possibly allows for the differing drapes of the fabrics to be more evident.</p><p>　　Figure 11 provides t

97、he 2D stiffness results measured in average bending lengths. Due to the inability of this method to capture the stiffness of fabrics in both warp and weft directions simultaneously, the results for each direction are pre

98、sented separately. A longer bending length suggests a stiffer fabric and hence a fabric with less drape, and vice versa. The finished acetate has the least drape due to its longest average bending lengths in both warp an