Objective Evaluation of Pucker in Three-Dimensional Garment Seams

Abstract

This paper considers the objective evaluation of seam puckering on three-dimensional garments. In order to simulate garment surfaces in three dimensions, seamed samples are supported on five different surfaces: plane (both with and without tilt) and curved (both convex and concave). A set of five samples having seams with varying degrees of pucker are prepared from each of five fabrics, and profiles parallel to the seams are measured using a laser distance scanner. These profiles are then filtered using a band-pass Chebyshev filter (where the "frequencies" are measured in cycles/m) to remove the highfrequency component contributed by the individual threads or fabric texture and the low-frequency component contributed by the shape of the sample. These band-pass filtered profiles are analyzed by calculating four statistical parameters and comparing these with the subjective AATCC grades. Excellent correlation is obtained for two of the parameters with the subjective pucker grades. Seam pucker is regarded as a matter of primary concern in garment manufacture. When the sewing parameters are not appropriate to the material selected, puckering occurs along the garment seams, and the aesthetic appeal of the garment deteriorates. The standard AATCC grading method [ I ] is well established but subjective, and therefore subject to error and variability between occasions and locations, particularly for patterned materials and darker shades. It is also time-consuming and involves several people.From the 50s to the 70s, several instruments [2, 3,8] using photo or displacement sensors were developed to measure the surface contours of seams. However, these instruments were not widely accepted due to problems with accuracy and reproducibility. With the development and wide application of computer and sensor technology, a number of new attempts were made in recent years. In 1992, Stylios and his co-workers [10, 11 used a CCD camera to capture the image of a seam,,and this information was then fed to a computer that could predict the grade or severity of seam pucker using an artificial intelligence algorithm. The CCD cameras were not entirely appropriate due to difficulties with patterned fabrics, and were replaced by laser scanners [9]. In the last few years, a number of other researchers [4, 5, 6] also did similar work in this field using laser scanners.Although considerable effort has been put into the objective evaluation of seam pucker, all the work to date has been on samples laid on a plane surface, which is not directly applicable to garments where few seams are completely flat. In this paper, we describe a method for extracting the pucker signal from seam profiles measured with a laser scanner, when the seamed samples are laid on both curved and tilted surfaces to simulate seams in three-dimensional garment surfaces. Four statistical parameters are derived from the pucker profiles and compared with the grades obtained by the AATCC method. Excellent agreement is demonstrated by a linear correlation with two of the paramete...

Objective Evaluation of Garment Seams Using 3D Laser Scanning Technology

“…5. With a pucker profile such as that shown in Figure 2, the logarithm of the variance, log (or2), of the height of the pucker profile was derived to provide a measure of seam pucker severity based on our previous work [3,4]. o~ is defined by the following equation: where z(i) is the height of the ith point on the pucker profile and N is the total number of points on the pucker profile.…”

“…Although there has been considerable work [2,5,6,7,12,13,14,15, 16] on the objective evaluation of seam pucker, most studies have involved isolated seam samples (i.e., seams sewn on strips of fabrics) laid on a plane surface; these cannot be directly applied to garment seams because garment surfaces are generally three-dimensional. In order to evaluate three-dimensional garment seams objectively, we have recently proposed using digital filters to extract pucker profiles by eliminating the signals representing the curvature of the garment surfaces and the fabric surface texture [3,4]. However, our previous work involved simulated garment seams (stripes of seams laid on different surfaces) and a 2D laser scanner.…”

Image-based evaluation of seam puckering appearance

“…This research shows that it is possible to predict and optimize the quality of seam pucker based on material properties and processing parameters. Fan 6 and Fan et al 7 used a commercial 3-D laser scanning system consisting of a laser scanning head, robot arm, computer, and some special software for data acquisition to scan the 3-D profile of garment seams. They used the 2-D filters to obtain the 3-D pucker profiles by removing the high-frequency components in the seam profiles as these could come from individual fabric threads or from noise.…”

Image analysis for seam-puckering evaluation

“…This surface structure has made 3D image analysis particularly suitable for seampuckering evaluation. 3,4 Fan et al 7 used a laser scanner and a band pass filter to extract the fabric surface profiles parallel to the seamline. The features computed from the profiles exhibited excellent correlations with the subjective puckering grades.…”