A computer vision-based system for automatic detection of misarranged warp yarns in yarn-dyed fabric. Part II: warp region segmentation

This series of studies aim to develop a computer vision-based system for automatic detection of misarranged color warp yarns to replace manpower and improve efficiency. Based on the warp yarn segmentation and fabric image stitching methods presented in Part I, this paper proposes a stepwise segmentation method of warp regions, as a core of the developed computer vision-based system, to detect the layout of color yarns for yarn-dyed fabrics automatically. The proposed framework consists of two main components: rough warp region segmentation and precise warp region merging which are realized by analyzing correlation coefficient of color histograms among segmented warp yarns and warp regions successively. The proposed method has been evaluated on 543 fabric images of four fabric samples consisting of 5533 warp regions, and experimental results show that the proposed method can realize the warp region segmentation in yarn dyed fabrics with the average accuracy of 99.47%.     

采用色纺纱图像的真实感色织物模拟

为解决色织物产品设计周期长,试织打样耗时费力的问题,提出一种采用色纺纱图像的真实感色织物的模拟方法。首先采集彩色纱线图像,运用阈值分割、形态学处理得到纱线主体,获取纱线主体的上、下边界和中心线,进而得到原始纱线图像的主体部分;接着根据椭圆模型和正弦曲线模型对纱线主体图像进行处理,得到纱线在二维织物表面中的形态;最后根据色纱循环和织物组织变换模型来改变经纬纱的覆盖关系,实现了真实感条纹型和格子型色织物的模拟。模拟结果表明：本文算法能够模拟不同种类色织物的织造过程,真实地反映织物的外观效果,且能够调整织物组织和色纱循环参数,提高了现有模拟算法的真实性和适应性。     

基于不同色纱组合的表里换层织物设计

概述了表里换层组织的基本结构和设计原理,分别设计了2种和3种色纱组合的表里换层织物,得出织物色彩数量与所用色纱数量的关系。结果表明:在改变经纬纱色纱排列次序的情况下,可以在表里换层织物表面形成不同深浅的几何色块,色块的数量随着所用色纱颜色的增加呈几何增长。将表里换层织物的组织变化与色纱变化相结合,可以形成独特的织物风格,为此类织物的设计提供了一种新的思路。     

Genetic algorithm‐based detection of the layout of color yarns

In this paper a novel method based on a genetic algorithm is proposed to recognize the layout of color yarns of yarn‐dyed fabric from the color pattern. The principle of a genetic algorithm is described first, and then the theories of roulette wheel selection method, crossover operation, and mutation operation are explained with the practical problem. Elitist selection is used to search for the correct result of the layout of color yarns. Some new chromosomes are added to the new generation in the genetic algorithm to avoid the local optimization. The repeat element of the layout of color yarns is then detected with period extraction. The repeat element of color pattern with the layout of color yarns is output together. Experiments on some color patterns recognized from actual yarn‐dyed fabrics, some color patterns simulated manually, and some color patterns including error color information of floats prove that the method proposed in this paper is effective for detecting the layout of color yarns from the color pattern of yarn‐dyed fabric, and it has a fault‐tolerance ability to some degree.     

Weave pattern recognition by measuring fiber orientation with Fourier transform

An effective method based on measuring the fiber orientation of yarn floats with two-dimensional Fourier transform (2-D FFT) is proposed to recognize the weave pattern of yarn-dyed fabric in the high-resolution image. The recognition process consists of four main steps: 1. High-resolution image reduction, 2.Fabric image skew correction, 3.Yarn floats localization, 4. Yarn floats classification. Firstly, the high-resolution image is reduced by the nearest interpolation algorithm. Secondly, the skew of the fabric image is corrected based on Hough transform. Thirdly, the yarn floats in the fabric image is localized by the yarns segmentation method based on the mathematical statistics of sub-images. Fourthly, the high-resolution image is corrected and its yarns are segmented successively according to the inspection information of the reduced image. The fiber orientations are detected by 2-D FFT, and the yarn floats are classified by k-means clustering algorithm. Experimental results and discussions demonstrate that, by measuring the fiber orientation of yarn floats, the proposed method is effective to recognize the yarn floats and the weave pattern for yarn-dyed, solid color, and gray fabrics.     

应用均值漂移的印花面料交互式换色方法

针对印花面料设计专业性要求高、面料打样耗时费力、对面料进行换色生产系列产品困难的问题,提出了一种基于纹理平滑与均值漂移(Mean-shift)的面料分色换色方法。采集并剪裁获得印花面料图像,采用相对总变差模型对面料图像进行纹理平滑,将平滑后的图像转换到CIE1976L^*a^*b^*色彩空间,然后利用Mean-shift聚类算法对面料图像进行分割,提取面料图像分色。通过交互式的方法调节分色,建立目标颜色与图像原始分色的映射关系,实现面料的分色换色,并采集不同的印花面料图像进行换色实验。结果表明:换色图像自然真实,面料纹理细节保留完整,可辅助印花面料设计,模拟印花面料的打样效果。     

基于卷积神经网络的机织物密度均匀性检测

针对目前基于机器视觉的机织物密度自动检测时织物检测视野小、精度低、品种适应性差的问题,提出一种基于多尺度卷积神经网络的检测方法。首先设计了一套离线图像采集系统连续采集织物图像,并建立一个包含详细织物参数的织物图像数据集;然后采用一种具有不同大小局部感受野的多尺度卷积神经网络适应不同大小的织物结构特征,定位纱线位置;最后利用霍夫变换及灰度投影方法处理网络模型所预测的纱线位置图,计算织物经纬密度,并对织物密度均匀性做出评价。结果表明:与其他方法相比,本文方法对于不同类型织物的经纬密度计算误差小于2%,检测精度更高,品种适应性更强。     

纯棉纱线合股数对织物性能的影响

为研究纯棉纱线合股数对织物力学性能和保形性的影响,采用线密度相同的单纱、双股线、3股线、4股线4种股线类型,以适当的经、纬密分别织制成平纹、斜纹、缎纹组织的织物。对织物的拉伸性能、撕裂强力、拉伸弹性、折皱回复性、悬垂性、弯曲性进行测试与分析。结果表明:3股线织物的断裂强力和撕裂强力较大,4股线织物的断裂伸长率较大,双股线织物的拉伸弹性回复率较大;平纹织物仅有双股线织物的折皱回复性优于单股线织物,斜纹和缎纹织物的折皱回复性随合股数的增加均变差;双股线平纹织物悬垂性相对较好,但均没有斜纹和缎纹织物的悬垂性好;单纱织物的弯曲性能比股线织物好。     

原液着色涤纶交织混色织物的颜色预测模型

为准确预测色纱交织后形成的空间混合颜色,对以往圆柱形纱线的织物几何结构模型进行了改进,并提出了织物几何结构模型中纱线压扁系数的计算方法;在现有有色纤维混合和色纱交织混合颜色预测模型的基础上,构建了一个带经验系数和中间函数的新模型用于原液着色涤纶交织混色织物的颜色预测,并提出采用牛顿迭代确定经验系数的方法。为检验模型的准确性和有效性,设计并织造了单组纬显色和双组纬显色,4种经纬密度组合以及32种组织结构的共288块原液着色涤纶长丝交织混色织物。研究结果表明,与以往固定形式的颜色预测模型相比,新模型的预测精度良好;在不同场合下,新模型可以调整经验系数,以适用于不同类型色纱交织混合颜色的更准确预测。     

增强水泥基用经编多轴向网格织物结构设计

针对增强水泥基用的经编多轴向网格织物进行结构设计,通过对增强水泥基复合材料所用的经编多轴向网格织物结构特点进行分析,以增强水泥基复合材料为出发点,从经编多轴向织物的网格尺寸和力学性能角度,对增强纱原料和规格的选择、增强纱的铺设方向和排纱方式以及捆绑结构的设计等方面进行了阐述,随后讨论,得出了关于纱线间隙和织物面密度的计算方法,最后设计和生产某种规格的经编四轴向网格织物,并对此织物的力学性能的验证,证明可以根据实践中水泥基复合材料的要求设计出相应的多轴向网格结构。     

Automatic inspection of yarn-dyed fabric density by mathematical statistics of sub-images

To inspect the yarn-dyed fabric density automatically, an effective image analysis method based on mathematical statistics of sub-images is proposed in this paper. This method consists of two main steps: rough measurement and precise measurement. The rough measurement is based on projection curve of the whole fabric image. The fabric image is converted into HSV model from RGB model firstly, and then the projection curve of value is gained directly. The number of yarns is obtained by counting the number of peaks in the curve roughly. The precise measurement is based on projection curves of the fabric sub-images. According to the roughly estimated yarn number, the whole fabric image is divided into a certain amount of sub-images and the projection method is applied to all the sub-images, respectively. The probability distribution map of peaks is obtained by processing the projection curves of all sub-images and the positions of the yarn center are located in the frequency curve generated from the map by mathematical statistics method. The number of peaks in the frequency curve is counted, and, therefore, the number of yarns is detected, and the density can be calculated precisely. The experimental results proved that the proposed method is effective for yarn-dyed fabrics and can satisfy the requirement for production practice.     

Analysis of mechanical and dimensional properties of the denim fabrics produced with double-core and core-spun weft yarns with different weft densities

Elastic structures are preferred for improving the elasticity and recovery properties of denim fabrics. With the intention of improving comfort during body movements in denim jeans, 10–35% elasticity is required. In this study, the effect of the composition of double-core (dual-core) and core-spun weft yarns and weft density on the mechanical properties of denim fabrics was investigated. For this purpose, different core materials were used in the production of elastic yarns. The sheath material was cotton for all yarn types. Yarn samples were spun with the same yarn count and twist coefficient. 100% cotton Ne 8/1 Ring slub warp yarns were used as warp yarn while double-core and core-spun yarns were used as weft yarns in weaving. The weaving process was performed with three different weft density values (16, 22, and 28 weft /cm) and the other production parameters were kept constant. Twill 3/1 woven fabrics were treated according to standard denim finishing procedures. After domestic washing processes, mechanical properties of the samples were tested and statistically evaluated.     

基于数字图像相关法的纱线和织物微观变形测量

为连续、精确地测量纱线和织物组织单元的拉伸变形量,结合数字图像相关法(DIC),提出一种微观尺度下的单轴拉伸变形测量方法。首先选取4种织物,从织物中抽取纱线、裁剪织物试样,并制作合理的散斑。然后对纱线和织物试样分别进行单轴拉伸实验并用相机记录整个过程。最后在DIC测试软件中添加引伸计,计算拉伸过程中单纱分段线应变以及织物单元结构在经、纬方向上的线应变。结果表明：单纱拉伸过程中,各处线应变增长速率不同;织物微小变形下内部结构呈非均匀变形,其中纬编针织物线圈变形不均匀性更为明显。基于DIC的微观变形测量方法揭示了纱线和织物生产、应用中的微量变形规律,为智能可穿戴电子纺织品的设计与应用提供指导。     

经向T结构预制体成型关键技术

为解决采用传统的“压扁-还原”的机织方法制备三维纺织品预制体成型的局限性问题,通过深入研究织造连续性、整体性T结构预制体的成型技术,采用以正交组织为基础,通过经纱接结绑定,以机织的方法织造梁高与底面厚度相等且一次成型的T结构三维机织物。对织造工艺进行优化改进,织造了接结纱跨越不同的纬纱列数的2种T结构三维机织物,测量了不同区域的尺寸。结果表明:1个完整组织循环中,为使跨越底面区域的浮线下沉且满足织造尺寸要求,底面区域纱线数约是总纱数的一半;当纬纱跨越2个纬纱列时,织物表面的平整度和成型性更好。     

异经提花条织物的生产实践

介绍了在喷气织机上织造两种不同经纱的提花织物的过程,具体论述了整经、浆纱织造时应注意的问题和采取的措施,有效解决了两种经纱的易绞现象,减少了经缩、横档等疵布,提高了产品质量和生产效率。     

An improved gray line profile method to inspect the warp–weft density of fabrics

Image processing has become a tremendous tool for various fields of applications as well as for textile manufacturing industry in recent years. Inspection of fabric density is one of the major issues for fabric manufacturers in textile industries. In this study, an image processing method comprising of linear and nonlinear techniques for automatic inspection of warp and weft yarn density of fabrics has been proposed. By avoiding common problems of linear filtering such as blurring and localization, anisotropic diffusion filtering has been applied as preprocessing operation to enhance the edge region/boundaries between adjacent yarns of the fabric images. We conjecture that given a skewed gray level image, the number of peaks in the gray line profile of the image is minimized if the image is rotated in such a way that the inter-spaces between yarns are aligned with the vertical axis. Gabor filter, an orientation-sensitive filter, is applied to the skewed image at that angle to boost the edges between inter-spaces. The number of warp and weft yarn density has been inspected by applying gray line profile method. Simulations have been done on a wide range of fabric image data set. The results have shown that nonlinear and steered filters made a contribution to the performance of the method. The number warp and weft yarn densities are determined with an accuracy rates above 90%.     

两面效应经编牛仔面料结构设计及其风格评价

针对靛蓝纱色牢度不高、贴近皮肤穿着易沾色及吸湿后皮肤有潮湿闷热感问题,同时为改善纬编牛仔面料尺寸稳定性差、线圈易脱散等缺点,以靛蓝纱和本白棉纱、异形截面和圆形截面涤纶黑丝与氨纶为原料,在四梳经编机上通过结构设计开发两面效应经编牛仔面料,使贴近皮肤的一面显露涤纶,另外一面显露棉纱。通过扫描电子显微镜对不同截面的涤纶表面进行表征,并对其拉伸性能进行比较;研究了面料的顶破性能、拉伸弹性回复性、吸湿速干性能,并利用PhabrOmeter 织物风格仪对风格特征进行评价。结果表明：靛蓝纱的强力高于本白棉纱,而毛羽指数降低了17.1%;异形截面涤纶显著提高了经编牛仔织物的滴水扩散速率、芯吸高度、手感和折痕回复性。     

Effect of structural parameters of porous yarns and fabric on air permeability and moisture transfer of double-face woven fabrics

In this study, effect of fabric structural parameters of double-face woven fabrics including kind of porous yarn namely micro-porous yarn and hollow yarn, hole size of hollow yarn, percentage of these yarns in double-face woven fabric structure, and finally weft density on air permeability and moisture transfer of woven fabrics was evaluated. These yarns were produced by using water-soluble continuous polyvinyl alcohol filaments as core part for hollow yarns and as doubling yarn in micro-porous yarn. Results revealed the effect of kind of porous yarn, hole size of hollow yarn, and weft density on air permeability and moisture transfer of woven fabrics. The percentage of porous yarns as weft did not show obvious trend. Analysis of variance was used to study the effect of these variables on air permeability and moisture transfer of double-face woven fabrics statistically.     

Yarn-dyed woven defect characterization and classification using combined features and support vector machine

In this paper, we present a novel defect evaluation method that uses combined features and modified support vector machine (SVM) classifiers to characterize and classify the defects of yarn-dyed fabrics. Yarn-dyed fabric images are preprocessed, and nine parameters are defined in the combined feature extractors. Based on binary and textural energy images for defect regions, yarn-dyed fabric defect features can be described, such as weft length, warp length, weft length to warp length ratio, perimeter, area, roundness, coarseness, contrast, and directionality. These parameters are also used as the inputs of optimized SVM classifiers to obtain overall defect classes in accordance with the Chinese National Standard of Yarn-dyed Pattern Fabrics (GB/T 22851 – 2009). The effectiveness of this evaluation method is tested by 180 selected defect images of yarn-dyed fabrics that have different patterns. The cross-validation tests on the yarn-dyed fabric defect classifications indicate that the defect categories of more than 91% of these diversified samples can be recognized correctly by using the SVM classification scheme. Furthermore, the extracted defect parameters provide useful information for textile and clothing manufacturing to grade yarn-dyed fabrics.     

The use of micro-computed tomography to determine the fabric cross-sectional area and stress

Accurate measurement of the cross-sectional area of porous fabric is quite important to determine fabric strength. Many models using yarns cross-section were built to calculate the strength of woven and knitted fabrics, and the spaces between fibres in the yarns were always considered to be part of the cross-sectional area. In our previous study, using micro-computed tomography (CT) and reconstructed three-dimensional (3D) images and based on the evaluation of the pore diameter distribution across the thickness direction, a new method was established to accurately determine the fabric ‘surface’ and fabric thickness. In this study, we used this method to calculate the cross-sectional area of fabrics and the resulting fabric stress. Using a plain weave and a flat knit fabrics showed that the cross-sectional areas in warp (or wale) and weft (or course) directions of the fabrics change regularly based on fabric structures and yarns. The fabric stress, when calculations were based on the least cross-sectional area, were greater than that calculated based on the other methods, such as the stress calculated based on the Peirce yarn model, where yarns are regarded as solid cylinders. Other research and properties related to the fabric thickness and cross-sectional areas will benefit from this accurate determination.