Objective Evaluation of Fabric Pilling Using Image Analysis Techniques

Abstract

For quality control of textiles, it is necessary to develop an objective method to evaluate fabric pilling, rather than subjective, to improve reliability and reproducibility. This paper presents a new method to extract pill features from fabric images. In this work, we apply a two-dimensional Gaussian fit theory to train a pill template using actual pill images for pill template matching, and determine a reasonable threshold for image segmentation using a histogram-fitting technique. We then extract five parameters to describe pill properties—pill number, mean area of pills. total area of pills. contrast. and density—and establish formulas to calculate objective grades with these parameters. The results show that a good correlation can be achieved between objective and subjective data.Pilling on fabrics is a well-known phenomenon, and such an unpleasant appearance can seriously compromise a fabric's acceptability. Pilling is an effect caused by wear and tear that considerably spoils the original appearance of a fabric. It begins with migration of fibers to the external parts of yarns, so that fuzz emerges on the fabric surface. Due to friction, this fuzz gets entangled, forming pills that remain suspended from the fabric by long fibers. Pills develop on a fabric surface in four stages: fuzz formation, entanglement, growth, and wearoff (Cooke, 1985) [2]. Fabric pilling is commonly tested in the laboratory by using specific machines to generate pills. These machines are usually supplied with a stan-. dard consisting of photographs of samples with different degree of pilling. Experts with long training and experience assign a degree of pilling by looking at the sample processed by the machine. However, a common drawback of these subjective methods based on estimations by experts is their inconsistency and the inaccuracy of the rating results.Great improvement can be achieved by introducing image analysis techniques, which have been widely used for characterizing and inspecting general textured materials and, in particular, textile materials [ 1. 7. 12]. Previous valuable works on pilling evaluations by digital image analysis have been reported in references 3, 4, 5, 7, 8, 9. 11. One-way of acquiring surface data from a fabric specimen is for a laser probe to measure surface height variation, as was pointed out by Ramgulam et al. [7]. Laser-scanning techniques can acquire 3D fabric depth images to avoid the difficulty of identifying pills on a patterned fabric. However, this kind of scanning process makes data acquisition much slower than camera capturing. since it needs an x -v stage to transport the sample mechanically. In this situation, a video camera. together with an effective algorithm to identify pills. is still of great research and practical value. Konda er al. (i988) [5] used a video camera and almost tangential illumination to capture samples. and Hsi et al. (1998) (4] found that a diffuse light source is much more suitable for pilling identification than collimated lighting. These groups i...

Image Analysis: Textile Industry in Indonesia

“…The collected data was tabulated by the content analysis, which focus on image analysis (Cohen, 1960;and Gottschalk, 2013): (Zhang & Bresee, 1995;and Wood, 1990);  Designs cloth (Zhang & Bresee, 1995;Xin, Hu, & Yan, 2002;and Rui-ling, Qing-Xiang, & Yu -he, 2005);  Fabric texture (Na & Pourdeyhimi, 1995);  Fabric structure (Kang, Choi, Kim, & Oh, 2001;Sakaguchi, Wen, Matsumoto, Toriumi & Kim, 2001;Jensen & Carstensen;and Escofet, Millán, & Rallo, 2001);  Strength of textile products (Xu, Pourdeyhimi, & Sobus, 1992;Kang, Choi, Kim, & Oh, 2001;and Jasper & Potlapalli, 1995); Price (Ting, 2015);  Quality fabrics (Sakaguchi, Wen, Matsumoto, Toriumi & Kim, 2001: Kim & Kang, 2005Xin, Hu & Baciu, 2010;Jensen & Carstensen, 2002;and Kang, Choi, Kim, & Oh, 2001);  Brand image (Bannister & Saunders, 1978).…”

“…In the same year, Xin, Hu, and Yan (2002) used the image analysis to extract the design's pill of the features of a clothes image, fabric-pilling evaluation, template matching pill, and they determine the reasonable limit of a histogram image of a segmentation fitting. There are five parameters that describe the amount of pills, as a nature-pill, pill region, the total area of pills, contrast and density of pills.…”

“… Designs cloth (Zhang & Bresee, 1995;Xin, Hu, & Yan, 2002;and Rui-ling, Qing-Xiang, & Yu -he, 2005);…”

Application of Wavelet Transform Method for Textile Material Feature Extraction

“…In addition, the fabric surface ruggedness may add further difficulty to fabric pilling feature extraction and assessment. Many researchers have tried to separate the pills from the background by image analysis techniques, such as pixel-based brightness threshold (Kim, S. & Park, 2006;Konda et al, 1988;Xin et al, 2002) and region-based template matching (Xin et al, 2002). However, these methods can not eliminate the influence of fabric texture.…”

Statistic learning-based defect detection for twill fabrics

“…Xin et al [11] used two-dimensional Gaussian template to detect pills via assuming them as elliptical or circular objects when illuminating fabrics at an oblique angle. Template matching methods are based on autocorrelation analysis by comparing texture image of samples to be checked with texture template produced from faultless fabrics.…”