Search results

This paper aims to determine the most practically applicable color-difference formula for yarn-dyed fabrics woven from warp and weft yarns in different color depths and to establish color-difference tolerance for perceptibility by evaluating yarn-dyed fabrics visually and instrumentally.

A total of 108 sample pairs were evaluated by a panel of 13 observers with perceptibility method under three typical light sources (A, D65 and cool white fluorescent). The data sets were statistically analyzed by the homogeneity of variance test (F-test), analysis of variance, standardized residual sum of squares and performance factor/3.

Light sources had a slight influence on the visual assessments of yarn-dyed fabrics. Among the eight color-difference formulae for measurements of yarn-dyed fabrics, CIEDE2000(2:1:1) outperformed all other tested formulae, and the color tolerance for the perceptibility of CIEDE2000(2:1:1) was 0.62. When the homochromy index (K) of warp and weft yarns of yarn-dyed fabric was lower than 1.25, the color difference based on ΔE*00(2:1:1) between the two samples was acceptable in terms of the color tolerance for perceptibility (i.e. 0.62).

The warp and weft yarns in different color depths could be woven in fabric with a relatively uniform color appearance.

This study could contribute to cost savings by reusing disqualified dyed yarns during the weaving manufacturing process.

The defect detection problem of color-patterned fabric is still a huge challenge due to the lack of manual defect labeling samples. Recently, many fabric defect detection algorithms based on feature engineering and deep learning have been proposed, but these methods have overdetection or miss-detection problems because they cannot adapt to the complex patterns of color-patterned fabrics. The purpose of this paper is to propose a defect detection framework based on unsupervised adversarial learning for image reconstruction to solve the above problems.

The proposed framework consists of three parts: a generator, a discriminator and an image postprocessing module. The generator is able to extract the features of the image and then reconstruct the image. The discriminator can supervise the generator to repair defects in the samples to improve the quality of image reconstruction. The multidifference image postprocessing module is used to obtain the final detection results of color-patterned fabric defects.

The proposed framework is compared with state-of-the-art methods on the public dataset YDFID-1(Yarn-Dyed Fabric Image Dataset-version1). The proposed framework is also validated on several classes in the MvTec AD dataset. The experimental results of various patterns/classes on YDFID-1 and MvTecAD demonstrate the effectiveness and superiority of this method in fabric defect detection.

It provides an automatic defect detection solution that is convenient for engineering applications for the inspection process of the color-patterned fabric manufacturing industry. A public dataset is provided for academia.

The purpose of this paper is introducing the image processing technology used for fabric analysis, which has the advantages of objective, digital and quick response.

This paper briefly describes the key process and module of some typical automatic recognition systems for fabric analysis presented by previous researchers; the related methods and algorithms used for the texture and pattern identification are also introduced.

Compared with the traditional subjective method, the image processing technology method has been proved to be rapid, accurate and reliable for quality control.

The future trends and limitations in the field of weave pattern recognition for woven fabrics have been summarized at the end of this paper.

The purpose of this paper is to present a novel method for minor fabric defects detection.

This paper proposes a PETM-CNN algorithm. PETM-CNN is designed based on self-similar estimation algorithm and Convolutional Neural Network. The PE (Patches Extractor) algorithm extracts patches that are possible to be defective patches to preprocess the fabric image. Then a TM-CNN (Triplet Metric CNN) method is designed to predict labels of the patches and the final label of the image. The TM-CNN can perform better than normal CNN.

This algorithm is superior to other algorithms on the data set of fabric images with minor defects. The proposed method achieves accurate classification of fabric images whether it has minor defects or not. The experimental results show that the approach is effective.

Traditional fabric defects detection is not effective as minor defects detection, so this paper develops a method of minor fabric images classification based on self-similar estimation and CNN. This paper offers the first investigation of minor fabric defects.

The purpose of this paper is to develop a 3D geometric model of three-pick terry fabric considering the actual design and structural features of the finished terry fabric.

The model has been developed using SolidWorks CAD system and the output file can be easily simulated in the ANSYS. Dimensions are acquired from the actual terry fabric measurement.

A 3D computational model – to be used for understanding the behaviour of terry fabric during actual use through the simulation in ANSYS.

Provides the way to study the yarn and fabric structure behaviour during use through simulation.

The research resulted a 3D geometrical model of very complex three-pick terry fabric for very first time for further analysis of terry fabric behaviour during use.

In recent years, rough set theory has evolved as one of the most promising classification techniques. One of the cardinal uses of rough set theory is its application for rule generation. The purpose of this paper is to propose a real-time fabric inspection technique. This work deals with the multi-class classification of fabric defects using rough set theory.

This technique focuses on the classification of fabric defects using the effective decision rules envisaged by rough set theory. In the proposed work, the six features of 50 images have been used for multiclass classification of fabric defects.

In this work, 40 images were used for generation of decision rules and 10 unseen images were used for validation out of which nine images are accurately predicted by the proposed technique.

The proposed method accurately identified 9 out of 10 testing defects. The obtained decision rules provide an insight about the classification method which ensures that the prediction accuracy can be improved further by framing more robust decision rules with the help of a large training data set. Thus, with the support of modern computational systems this method is potent in getting recognition from the textile industry as a real-time classification technique.

Titanium(IV) oxide nanoparticles (TiO₂ NP) were deposited to cotton denim fabrics using a self-crosslinking acrylate – a polymer dispersion to extend the lifetime of the products. This study aims to determine the optimum conditions to increase abrasion resistance, to provide self-cleaning properties of denim fabrics and to examine the effects of these applications on other physical properties.

The denim samples were first treated with nonionic surfactant to increase their wettability. Three different amounts of the polymer dispersion and two different pH levels were selected for the experimental design. The finishing process was applied to the fabrics with pad-dry-cure method.

The presence of the coatings and the adhesion of TiO₂ NPs to the surfaces were confirmed by scanning electron microscope and Fourier transform infrared spectroscopy analysis. It was ascertained that the most appropriate self-crosslinking acrylate amount and ambient pH level is 10 mL and “2”, respectively, for providing increased abrasion resistance (2,78%) and enhanced self-cleaning properties (363,4%) in the denim samples. The coating reduced the air permeability and softness of the denim samples. Differential scanning calorimetry and thermogravimetry analysis results showed that the treatments increased the crystallization temperatures and melting enthalpy values of the denim samples. Based on the thermal test results, it is clear that mass loss of the denim samples at 370°C decreased as the amount of self-crosslinking acrylate increased (at pH 3).

This study helped us to find out optimum amount of self-crosslinking acrylate and proper pH level for enhanced self-cleaning and abrasion strength on denim fabrics. With this finishing process, an environmentally friendly and long-life denim fabric was designed.

The objective of the research is to develop implementation strategies for producers at the fashion apparel supply chain upstream, in order to move towards a more coordinated, streamlined and responsive process.

Qualitative action research was conducted using non‐participatory observations on sampled producers, following a literature review on the design process and mass customization.

Main activities with contributing factors that funnel in and out of this crucial junction are mapped and broken down into a series of processes that involve producers' selection and customers' choice, where decisions are currently made based on informal correlation of supply push and demand pull, months ahead of end‐users' (“customers” hereon) real demand. Key “integrated decision points” where customers' input is identified and can be introduced into the outbound supply chain.

This conceptual model offers the possibilities for implementing collaborative mass customization with reduced risk for producers and increased satisfaction for customers. However, producers' resistance to change from existing work methods may present potential obstacles. Further work is to be done on collecting, utilizing, and transforming customers' data in order to inform the total design process effectively and comprehensively.

The results of the “integrated decision pulse point map” proposed by this paper provide a threshold to the benefits of mass customization at the heart of the fashion system.

The purpose of this paper is to propose a new recognition algorithm on quadratic local extremum for fabric density recognition.

The density wave is established to correctly detect density by searching local extremes. The gray wave of each line in fabric image is extracted first. The derivation of gray wave is calculated, extreme waves including all true extreme values and false extreme values are obtained. And then the second derivative of extreme waves are calculated and the result makes local correction. The density wave, which can represent position and quantity of yarn and interstice, is established. According to the resolution and size parameters of image, the function of density with density wave statistics is given.

The experiment and analysis proved that the method proposed can detect fabric density simply and successfully with less calculation and no image preprocessing.

The algorithm provides practical guidelines for fabric density detection and provides a new thought for fabric characteristic identification. Future work could be focused on the development of methods for the automatic algorithm with color fabrics.

The algorithm based on quadratic local extremum presented in this paper is a new method to successfully detect fabric density and can be applied to the recognition for other categories of clothing fabrics and images.

Visual quality control on raw textile fabrics is a vital process in weaving factories to ensure their exterior quality (visual defects or imperfection) satisfying customer requirements. Commonly, this critical process is manually conducted by human inspectors, which can hardly provide a fast and reliable inspection results due to fatigue and subjective errors. To meet modern production needs, it is highly demanded to develop an automated defect inspection system by replacing human eyes with computer vision.

As a structural texture, fabric textures can be effectively represented by a linearly summation of basic elements (dictionary). To create a robust representation of a fabric texture in an unsupervised manner, a smooth constraint is imposed on dictionary learning model. Such representation is robust to defects when using it to recover a defective image. Thus an abnormal map (likelihood of defective regions) can be computed by measuring similarity between recovered version and itself. Finally, the total variation (TV) based model is built to segment defects on the abnormal map.

Different from traditional dictionary learning method, a smooth constraint is introduced in dictionary learning that not only able to create a robust representation for fabric textures but also avoid the selection of dictionary size. In addition, a TV based model is designed according to defects' characteristics. The experimental results demonstrate that (1) the dictionary with smooth constraint can generate a more robust representation of fabric textures compared to traditional dictionary; (2) the TV based model can achieve a robust and good segmentation result.

The major originality of the proposed method are: (1) Dictionary size can be set as a constant instead of selecting it empirically; (2) The total variation based model is built, which can enhance less salient defects, improving segmentation performance significantly.