Search

The application of content-based image retrieval method aims at retrieving similar fabric images and obtaining the existing process parameters to guide production. The process of sample analysis, trial weaving, and proofing can be eliminated in sample imitation production to give full play to the advantages of historical production experience and improve the core competitiveness of enterprises. By investigating and analyzing the applications of content-based image retrieval method technology in fabric retrieval, this article provides a detailed classification and summary of the existing fabric retrieval methods using content-based image retrieval method from six aspects: image preprocessing, feature extraction, similarity measurement, retrieval strategy, dataset construction, and evaluation metrics in the common framework of content-based image retrieval method. The advantages and disadvantages of different methods are analyzed and compared. Finally, the urgent problems and future research directions of fabric image retrieval are discussed, providing ideas for scholars to further study the retrieval methods. Taking fabric as the medium, this article combs the industrial application research and development process of content-based image retrieval method technology, which is helpful to understand the application examples of computer technology and provide research ideas for the application of different computer technologies in the textile industry.

Compared with solid-colored fabrics, the textures in yarn-dyed fabric images are more complex, making the task of defect detection more challenging. To achieve efficient detection, this study proposes an automatic detection framework for dyed fabric defects. The proposed framework consists of a hardware system and a detection algorithm. For efficient and high-quality acquisition of fabric images, an image acquisition assembly equipped with three sets of light sources and a mirror was developed. In addition, a defect detection algorithm based on Fourier convolution and a convolutional autoencoder is proposed. Abandoning the common way of adding noise, this paper proposes to generate image pairs for training using a random masking method in the training phase. In the autoencoder, some traditional convolutional layers are replaced with Fourier convolutional layers. Ablation experiments verify the effectiveness of the mask generation method and Fourier convolution. Compared with other defect detection methods, the proposed method achieves the best performance, which verifies the superiority of the method. The maximum detection speed of the developed system can reach 41 meters per minute, which can meet real-time requirements.

Extracellular proteases of haloarchaea are adapted to function at high concentrations of NaCl, and could find useful applications in industrial or biotechnological processes where hypersaline conditions are desired. The diversity of extracellular proteases produced by haloarchaea is largely unknown, although the genomes of many species have been sequenced and are publicly available. In this study, a gene encoding the extracellular protease Hly176B from the haloarchaeon Haloarchaeobius sp. FL176 was cloned and expressed in Escherichia coli. The catalytic triad Asp-His-Ser was confirmed via site-directed mutagenesis, indicating that Hly176B belongs to the class of serine proteases (halolysin). Unlike previously reported extracellular proteases from haloarchaea, Hly176B remained active in almost salt-free solution for a relatively long time. In addition, Hly176B displayed good tolerance to some metal ions, surfactants and organic solvents, and exerts its highest enzyme activities at 40 °C, pH 8.0 and 0.5 M NaCl. Therefore, this study enriches our knowledge of extracellular proteases, and also expands their applications in various industrial uses.

Purpose The paper aims to propose a novel method based on deep sparse convolutional neural network (CNN) for clothing recognition. A CNN based on inception module is applied to bridge pixel-level features and high-level category labels. In order to improve the robustness accuracy of the network, six transformation methods are used to preprocess images. To avoid representational bottlenecks, small-sized convolution kernels are adopted in the network. This method first pretrains the network on ImageNet and then fine-tune the model in clothing data set.Design/methodology/approach The paper opts for an exploratory study by using the control variable comparison method. To verify the rationality of the network structure, lateral contrast experiments with common network structures such as VGG, GoogLeNet and AlexNet, and longitudinal contrast tests with different structures from one another are performed on the created clothing image data sets. The indicators of comparison include accuracy, average recall, average precise and F-1 score.Findings Compared with common methods, the experimental results show that the proposed network has better performance on clothing recognition. It is also can be found that larger input size can effectively improve accuracy. By analyzing the output structure of the model, the model learns a certain “rules” of human recognition clothing.Originality/value Clothing analysis and recognition is a meaningful issue, due to its potential values in many areas, including fashion design, e-commerce and retrieval system. Meanwhile, it is challenging because of the diversity of clothing appearance and background. Thus, this paper raises a network based on deep sparse CNN to realize clothing recognition.

Fine-texture Image Retrieval, a special case in Content Based Image Retrieval, has high potential application value in many fields. However, there are very few researches focus on this special case. Due to the difference between fine-texture images and common images, general retrieval methods for common images are difficult to apply to fine-texture image retrieval. It is also a challenging issue with several obstacles: variety and complexity of appearance, as well as high requirements for retrieval accuracy. To address this issue, this study proposes a novel approach, called deep multi-view hashing (DMVH), to learn enhanced hash codes for efficient fine-texture image retrieval. We propose to use the first few layers of a deep convolutional neural network for fine-texture image presentation. To avoid overfitting, we employ L0Linear layer instead of the commonly used Linear layer in all the fully connected layers. Then we introduce a pairwise quantified similarity computed on the semantic labels. And the learning process of the model is guided by multiple labels of images. This study takes fabric as an example, and builds a dataset called MFT-fabric-v1, to validate the effectiveness of the DMVH method. On this data set, the proposed model achieved a performance with a mAP value of o.8735. The area under the pr curve is larger than other comparison methods. The experimental results demonstrate that the proposed method outperforms the state-of-the-art on the MFT-fabric-v1 dataset.

Under the production mode of small-batch and multi-item, the recognition of yarn-dyed fabric patterns is a crucial task in the textile industry. In this article, an automatic recognition system based on pixel-level features is proposed to recognize the density, the weave pattern, and the color pattern. In this system, the fabric images are captured by a scanner. First, a method based on the Hough transform is used to correct the skew of the yarns, including warp and weft. Second, the yarns and nodes are located in the enhanced images with a brightness-projection method. The density can be calculated by using the results. Then, the type of each node is identified based on the boundary information. We can obtain the weave pattern after knowing the type of each node. Finally, the fuzzy C-means algorithm is used to determine the color of each node, and thus we obtain the color pattern of the yarn-dyed fabric. Experimental results demonstrate that the proposed recognition system is effective for detecting the structural parameters of yarn-dyed fabric.

Fabric surface flaw inspection is essential for textile quality control, and it is demanding to replace human inspectors with the automatic machine vision-based flaw inspection system. To alleviate the time-consuming problem of sparse coding in detecting phase, this work presents a real time fabric flaw inspection method by using grouped sparse dictionary. Firstly, the over-complete sparse dictionary is learned from normal fabric images; Secondly, the learned sparse dictionary is grouped into several sub-dictionaries by evaluating reconstruction error. Finally, the grouped dictionary is used to represent image and identify flaw regions as they cannot be represented well, leading to large reconstruction error. In addition, a non-maximum suppression algorithm is also proposed to reduce false inspection further. Experiments on various fabric flaws and real-time implementation on the proposed vision-based hardware system are conducted to evaluate the performance of proposed method. In comparison with other dictionary learning methods, the experimental results demonstrate that the proposed method can reduce the running time significantly and achieve a decent performance, which is capable of meeting the real-time inspection requirement without compromising inspection accuracy.

Fabric shape retention is a crucial property that impacts the durability and usability of fabric and which needs a convenient and accurate evaluation method. In the previous research, the automated crease recovery testing method was used to obtain fabric crease recovery information and evaluate the property of shape retention. Based on the previous research, an orthogonal test method was adopted to investigate the effect of different test parameters on the stability of shape retention detection in this paper. First, three factors, that of sample size, pressure, and pressure time, and three different levels of each factor were determined by the L9(34) orthogonal test table. Next, the fabric shape retention indexes were detected by nine different test schemes, and the comprehensive score of shape retention index standard deviation was obtained as the evaluation criterion of the orthogonal test results. Finally, the optimal test scheme was determined by visual analysis and variance analysis. The results show that the sample size has a certain effect on the stability of shape retention indexes, while the pressure time and the pressure have no significant influence. The optimal test scheme is that the sample size is 30 cm × 30 cm, the pressure time is 60 s, and the pressure is 40 N. The test result measured by the optimal test parameters reveals excellent stability when the vertex angle standard deviation is 2.0°, the height standard deviation is 0.06 cm, and the shape retention area standard deviation is 0.16 cm2. This paper provides an experimental basis for improving the accuracy of fabric shape retention evaluation method.

For sample reproduction, texture and color are both significant when the consumer has no specific or individual demands or cannot describe the requirements clearly. In this paper, an effective method based on aggregated convolutional descriptors and approximate nearest neighbors search was proposed to combine the texture and color feature for wool fabric retrieval. Aggregated convolutional descriptors from different layers were combined to characterize the wool fabric image. The approximate nearest neighbors search method Annoy was adopted for similarity measurement to balance the trade-off between the search performance and the elapsed time. A wool fabric image database containing 82,073 images was built to demonstrate the efficacy of the proposed method. Different feature extraction and similarity measurement methods were compared with the proposed method. Experimental results indicate that the proposed method can combine the texture and color feature, being effective and superior for image retrieval of wool fabric. The proposed scheme can provide references for the worker in the factory, saving a great deal of labor and material resources.

Due to the potential value in many areas, such as e-commerce and inventory management, fabric image retrieval, which is a special case of content-based image retrieval, has recently become a research hotspot. As a major category of textile fabrics, patterned fabrics have a diverse and complex appearance, making the retrieval task more challenging. To address this situation, this paper proposes a novel approach for patterned fabric based on the non-subsampled contourlet transform (NSCT) feature descriptor and relevance feedback technique. To integrate the color information into the NSCT feature descriptor, we extract the feature of patterned fabric images in HSV color space. An outlier rejection-based parametric relevance feedback algorithm is employed to adjust the similarity matrix to improve the retrieval results. The experimental results not only show the effectiveness of the proposed approach but also demonstrate that it can significantly improve the performance of the retrieval system compared to other state-of-the-art algorithms.

Yarn evenness and hairiness are the appearance characteristics of yarn, which affect textile processing and product quality. To evaluate yarn appearance economically and effectively, an image-processing method is proposed in this paper to analyze yarn appearance on a blackboard. Firstly, an image of a yarn blackboard is captured by the scanner. Then, the yarn core and hairy fibers are segmented from the captured image with image-processing algorithms. The coefficients of variation of the yarn diameter ( CVbd) and the hairiness index ( M) are respectively calculated based on the information about the yarn core and hairy fibers in the image. Finally, the results of the proposed method are compared with those from the Uster Tester. The experimental results demonstrate that yarn appearance can be objectively evaluated using yarn blackboard images. The test results of different yarn blackboards made from the same yarn are stable and consistent. The correlation coefficient between the proposed method and the Uster Tester is 0.98, which proves that the H value can be accurately predicted by the hairiness prediction model. A hairiness prediction model built by the M value is also proven to be accurate when used to predict the corresponding value of the Uster Tester. Compared with the existing yarn evenness and hairiness test methods, the proposed method is more economical and practical.

Texture provides important visual information for object identification, and texture representation is still a challenging problem in texture analysis. Fabric texture is a typical structural textur...

Computer-based fabric segmentation has recently increased significantly in diverse fields of which textiles design and engineering is no exception. Extracting appropriate information from printed f...

Computer color matching can improve production efficiency and reduce costs in color spun. However, in practice the computer color matching success rate for pre-colored fiber blends has not been good, leading to customers being unsatisfied with the accuracy of the color matching results. Aiming to improve the accuracy, a hybrid of least squares and grid search method has been proposed for spectrophotometric color matching of pre-colored fiber blend based on the improved Kubelka–Munk (K-M) double-constant theory. Two-primary, three-primary, four-primary, and five-primary pre-colored cotton fiber blends were prepared as standard samples to evaluate the color matching accuracy of the proposed method. Compared with the least squares method and the grid search method, the proposed method achieved better color matching effects and greatly shortened the calculation time, respectively. For 42 pre-colored fiber blends, the average color difference between the predicted results obtained by the proposed method, least squares method, and grid search method and the spectrophotometer measurements were respectively 0.29, 0.53, and 0.36 CIE2000 units. The experimental results indicated that the proposed method could predict the formulation of standard samples quickly and effectively, and that it was superior to other methods in providing satisfactory color matching results for the enterprises.

The accuracy of color prediction results for pre-colored fiber blends is critical in the textile industry. In this paper, we attempt to investigate a feasible method for predicting the color of pre-colored fibers blends. Five pre-colored cotton fibers were divided into two groups, one for achromatic primaries (white and black) and one for chromatic primaries (red, blue, and yellow). Their respective absorption coefficient (K) and scattering coefficient (S) were calculated by the least squares method from the prepared fiber blends samples. The color information of the 34 test blending samples including two-primary and three-primary was predicted by the improved Kubelka-Munk (K-M) double-constant theory. Comparing with the measurement results, the minimum and maximum DE00 color differences were 0.215 and 1.890. The variance of color difference for two-primary samples was 0.128 and for three-primary samples was 0.154, both were smaller than that obtained by the K-M theory relative value method, the Stearns-Noechel (S-N) model, revised S-N models, and the Friele model. The results show that the improved K-M double-constant theory can be used to better predict the color blending effect of pre-colored fibers.

The fabric of colored spun yarn has ever-changing appearances and styles with different fancy yarns. The fabric image is commonly designed by the designer using the software, which needs complex user interactions and difficult image segmentation. In this paper, a modified color transfer method was proposed to generate the fabric appearance of colored spun yarn. Given the color card as the target image, the style fabric image was matched as the reference image based on the dominant luminance. After transferring the two images to lαβ color space, Wavelet transform and luminance sampling were utilized to filter the redundant high-frequency information and select the representative pixels, respectively. Then, the chromatic channels were transferred based on the best matched luminance and the neighborhood relation. Finally, the image after color transfer was reconstructed by wavelet reconstruction. The proposed reference image matching maintained the result to be the ground truth. For the samples selected, the combined methods based on wavelet transform and luminance sampling improved the efficiency and performance of the proposed scheme. Experiments were conducted on different fabrics with different colors and styles. Experiments demonstrated the validity and superiority of the proposed method, which can provide referential assistance for the designer and save considerable amounts of labor.

The outbreak of coronavirus has led to an increase in the demand for facemasks globally. Unavailability of appropriate polypropylene non-woven fabrics face masks as a result of inadequate supply to satisfy the growing population has brought about the manufacturing of locally fabrics masks to augment or substitute standard medical class facemasks. The study aims at analyzing airflow of these locally manufactured fabrics to determine possible means of transmitting the virus as well as establish comfort of the user of these masks. Standard polypropylene non-woven, woven and knitted fabrics were considered for the study. Air permeability test was conducted on these fabrics using Frazier Air permeability tester. Depending on the property significant variation in the textile fabrics, polypropylene non-woven is widely accepted for facial masks. Nevertheless, this study illustrates that woven and knitted fabrics have more open structures, which allow a high rate of air penetration and so may require two or three layers to prevent antimicrobial or antiviral potential.

Image processing of digital images is one of the essential categories of image transformation in the theory and practice of digital pattern analysis and computer vision. Automated pattern recognition systems are much needed in the textile industry more importantly when the quality control of products is a significant problem. The printed fabric pattern segmentation procedure is carried out since human interaction proves to be unsatisfactory and costly. Hence, to reduce the cost and wastage of time, automatic segmentation and pattern recognition are required. Several robust and efficient segmentation algorithms are established for pattern recognition. In this paper, different automated methods are presented to segregate printed patterns from textiles fabric. This has become necessary because quality product devoid of any disturbances is the ultimate aim of the textile printing industry.

In order to establish an objective, stable and efficient wrinkle evaluation system for fabric wrinkle evaluation, a method based on 2-D Gabor transform was proposed. Among this system, the directions of Gabor filter were determined according to the range of amplitude response. Then a set of Gabor filters were obtained by selecting and optimizing the central frequency, the half peak bandwidth and the shape factor of the Gaussian surface. After Gabor transform by such filter bank, the amplitudes of different response spectrums were extracted, constructing a multi-dimensional feature vector. Finally, the feature vectors of the fabric image samples, whose wrinkle degrees were evaluated manually in advance, were extracted and used to train a support vector machine (SVM), which achieved 81.82 % evaluation accuracy on the 345 samples. The trained SVM was applied to evaluate the wrinkle degree of the fabric samples acquired in different illumination directions, and verified the stability of the proposed method to illumination environment. Compared with the existing method, the proposed method has higher classification accuracy. The comparison results indicate the Gabor amplitude feature proposed by this research has a high correlation with the fabric wrinkle grades.

Weaving enterprises are faced with problems of small batches and many varieties, which leads to difficulties in manual scheduling during the production process, resulting in more delays in delivery. Therefore, an automatic scheduling method for the weaving process is proposed in this paper. Firstly, a weaving production scheduling model is established based on the conditions and requirements during actual production. By introducing flexible model constraints, the applicability of the model has been greatly expanded. Then, an improved ant colony algorithm is proposed to solve the model. To address the problem of the traditional ant colony algorithm that the optimizing process usually traps into local optimum, the proposed algorithm adopts an iterative threshold and the maximum and minimum ant colony system. In addition, the initial path pheromone distribution is formed according to the urgency of the order to balance each objective. Finally, the simulation experiments confirm that the proposed method achieves superior performance compared with manual scheduling and other automatic methods. The proposed method shows a certain guiding significance for weaving scheduling in practice.

Objective fabric smoothness appearance evaluation plays an important role in the textile and apparel industry. In most previous studies, objective fabric smoothness appearance evaluation is defined as a general pattern classification problem. However, the labels in this problem exhibit a natural ordering. Nominal classification ignores the ordinal information, which may cause overfitting in model training. In addition, for the existence of subjective errors, measurement errors, manual errors, etc., the labels in the data might be noisy, which has been rarely discussed previously. This paper proposes an ordinal classification framework based on label noise estimation (OCF-LNE) to objectively evaluate the fabric smoothness appearance degree, which takes the ordinal information and noise of the label in the training data into consideration. The OCF-LNE uses the basic classifier in pre-training as a label noise estimator, and uses the estimated label noise to adjust the labels in further training. The adjusted labels can introduce the ordinal constrain implicitly and reduce the negative impact of label noise in model training. Within a 10 × 10 nested cross-validation, the proposed OCF-LNE achieves 82.86%, 94.29%, and 100% average accuracies under errors of 0, 0.5, and 1 degree, respectively. Experiments on different fabric image features and basic classification models verify the effectiveness of the OCF-LNE. In addition, the proposed method outperforms the state-of-the-art methods for fabric smoothness evaluation and ordinal classification. Promisingly, the OCF-LNE can provide novel ideas for image-based fabric smoothness evaluation.

The layout of colored yarns in yarn-dyed fabrics is a significant part of designing and production in the textile industry, which is still analyzed manually at present. Existing methods based on image processing have some limitations in accuracy and stability. Therefore, an automatic method is proposed to recognize the layout of colored yarns and some other basic fabric structure parameters: the fabric density and weave pattern. First, a large dataset with fabric structure parameters is constructed. The fabric images are captured by a wireless portable device. Then the yarns and floats are accurately located using a novel multi-task and multi-scale convolutional neural network. Finally, a density-based color clustering algorithm is proposed to recognize the layout of colored yarns. The results of extensive experiments show that the proposed method can automatically identify the basic structure parameters with high effectiveness and robustness.

The recognition of woven fabric pattern is a crucial task for mass manufacturing and quality control in the textile industry. Traditional methods based on image processing have some limitations on accuracy and stability. In this paper, an automatic method is proposed to jointly realize yarn location and weave pattern recognition. First, a new big fabric dataset is established by a portable wireless device. The dataset contains wide kinds of fabrics and detailed fabric structure parameters. Then, a novel multi-task and multi-scale convolutional neural network (MTMSnet) is proposed to predict the location maps of yarns and floats. By adopting the multi-task structure, the MTMSnet can better learn the related features between yarns and floats. Finally, the weave pattern and basic weave repeat are recognized by combining the yarn and float location maps. Extensive experimental results on various kinds of fabrics indicate that the proposed method achieves high accuracy and quality in weave pattern recognition.

With the advancing consumption level, it is difficult to shorten the product cycle and meet consumer’s demands during the design process of yarn-dyed plaid fabric. In order to extract consumers’ pr...

This series of studies aims to develop a computer vision-based system for automatic detection of misarranged color warp yarns. This paper proposes a yarn layout proofing strategy, integrating with ...

Fabric shape retention is one of the most important attributes of fabrics that can influence the quality of the end use product. In this paper, we present a computer vision-based method to analyze the sequential images, which records the dynamic change of a deformed fabric, to model the recovery process, and extract the features of the recovery curve to characterize the shape retention after the deformation. Image processing and the perceptual hash algorithm were used to convert the measurements of a fabric shape variable at different times into Hamming distance points. The recovery function of the fabric shape was formed by fitting the Hamming distance points with exponential function, and three new shape retention indexes, that is, the average slope, the abscissa of the inflation point, and the radius of curvature at the inflation point, were defined based on the recovery function. The experiment showed that the shape retention of 12 fabric samples after deformation could be effectively distinguished by the new indexes. This paper also discussed the relationships between the new indexes and the transitional measurements indicating the fabric shape retention.

Deep learning-based defect inspection has gained popularity in recent years. The dataset requirements for the supervised learning-based method are currently high, but the types of defects are numerous and difficult to gather. This work proposes a local image reconstruction-based unsupervised fabric defect segmentation method to address this problem. Cyclic structures make up the normal portion of the fabric image, whereas the defects are anomalous and minor in comparison. As a result, the defect will be recreated as a normal texture utilizing the information from its surrounding areas, and the defect information will be preserved in the residual image. By masking the same area with various shapes, different reconstruction outcomes and residual images can be achieved. The signal of the defect will be amplified and the noise will be decreased due to the random distribution when the generated residual pictures are fused, which can effectively identify the defect from the noise and lower the false detection rate. On the denim fabric dataset, the proposed unsupervised method can achieve high precision fabric defect segmentation, with the defect detection rate and detection precision reaching at least 85% and 89%, respectively, with high efficiency (approximately 60 m/min inspection speed), outperforming other fabric defect segmentation methods.

In the textile and apparel industry, it remains a challenging task to evaluate the fabric smoothness objectively. In existing studies, the objective fabric smoothness assessment is defined as a typical image classification problem. However, the fabric smoothness labels contain sequence information, and the problem shall be defined as an ordinal classification problem. This article presents an effective method including an image preprocessing algorithm, a compact convolutional neural network(CNN) model, and a label smoothing process. Compared with the commonly used CNN frameworks, the proposed compact CNN model is more suitable for this small-sample and low-abstraction problem. The image processing algorithm can improve the model’s illumination adaptability, and the label smoothing process can modify the model to satisfy the ordinal classification problems better. In the experiments, the method is tested on a fabric image set including 385 graded fabric specimens. Within a 10-fold cross validation, the proposed method achieves 84.00%, 95.38%, and 100% average accuracies under errors of 0 degree, 0.5 degree, and 1 degree respectively. Implementation discussions on preprocessing and label smoothing verify their effectiveness in improving model performance in assessment accuracies and illumination stability. The proposed method outperforms the state-of-the-art methods for fabric smoothness assessment and a series of widely used deep learning methods. Promisingly, the proposed method can provide novel research ideas for the image-based fabric smoothness assessment.

Due to the significant potential values in commercial and social applications, clothing image recognition has recently become a research hotspot, among which clothing attribute recognition is an important content. However, the large variations in the appearance and style of clothing and the image’s complex forming conditions make the task challenging. Moreover, a generic treatment with deep convolutional neural networks cannot provide an ideal solution. Instead of using CNNs for classification, we proposed a novel approach based RCNN framework for the recognition task. Firstly, we apply the modified selective search algorithm to extract the region proposal. Then, the Inception-ResNet V1 model with L-Softmax is employed to represent images and identify their categories. After Soft-NMS, we use a simple neural network to correct the boundary of region box. To evaluate the performance of the framework, a dataset including about 100,000 shirt images was built. The experimental result show that our proposed framework achieved promising overall labeling rate, precision and recall of 87.77%, 73.59% and 83.84%. In addition, comparative experiments demonstrate the superiority of the proposed framework.

In the textile and apparel industry, it remains a challenging task to evaluate the fabric smoothness appearance objectively. In existing studies, with computer vision technology, researchers use the hand-crafted image features and deep convolutional neural network (CNN) based image features to describe the fabric smoothness appearance. This paper presents an image classification framework to evaluate the fabric smoothness appearance degree. The framework contains a feature fusion module to fuse the hand-crafted and CNN features to take both advantages of them. The framework uses the multi-scale spatial masking model and a pre-trained CNN to extract hand-crafted and CNN features of fabric images respectively. In addition, a mislabeled sample filtering module is set in the framework, which helps to avoid the negative impact of mislabeled samples in training. In the experiments, the proposed framework achieves 85.2%, 96.1%, and 100% average evaluation accuracies under errors of 0 degree, 0.5 degree, and 1 degree respectively. The experiments on the feature fusion and mislabeled sample filtering verified their effectiveness in improving the evaluation accuracies and the label noise robustness. The proposed method outperforms the state-of-the-art methods for fabric smoothness assessment. Promisingly, this paper can provide novel research ideas for image-based fabric smoothness assessment and other similar tasks.

Color is difficult to distinguish by human vision and is described by keywords, resulting in low efficiency of wool fabric retrieval in factories at present. To obtain the process sheets of existing products and reduce the work of color measurement in sample analysis, this paper proposes an effective method based on dominant colors (DCs) and color moments (CMs) for wool fabric image retrieval. Firstly, the image was scaled to reduce computational time. Then, the hue, saturation, value color space was divided into 128 parts by the fast color quantization algorithm to extract the DCs of the image. Meanwhile, the CMs based on image partition were calculated in CIE L* a* b* color space to describe the spatial color information. Subsequently, different similarity measure methods were carried out based on the DC feature and CM feature. Finally, experiments were conducted on a wool fabric image database with 20,000 images for parameter optimization and verification. The average precision and recall were up to 87% and 44%, respectively. Experimental results show that the proposed scheme can retrieve images with the same or similar colors quickly and effectively and it outperformed other methods, providing referential assistance for the factory worker when retrieving wool fabrics.

Fabric smoothness appearance assessment plays an important role in the textile and apparel industry. To evaluate fabric smoothness objectively, different methods have been proposed based on computer vision technology. To further improve the performance and promote the application of the assessment methods, this paper reports a hybrid computer vision system for objective assessment of fabric smoothness appearance with an ensemble classifier to integrate the advantages of the different image feature sets, which are extracted based on different image processing technologies. The image acquisition environment is established in this system with the selection of illumination parameters—intensity, position angle and altitudinal angle—by a designed strategy. The main steps of the strategy include determination of priority by information gain analysis and parameter selection by classifier performance analysis. The support vector machine classifiers trained by each feature sets are grouped into an ensemble by a self-adapting weighted voting method and the redundant feature sets are eliminated based on the weights of the feature sets. The final result shows evaluation accuracies with 82.86% under 0-degree error, 97.14% under 0.5-degree error and 100% under 1-degree error, which outperforms the other methods in the same environment and verifies the applicability of the proposed system.

This article presents a computer vision method for measuring the geometrical parameters of slub yarn based on yarn sequence images captured from a moving slub yarn. An image segmentation method proposed by our earlier work was applied to segment sequence slub yarn images to obtain overlapping diameter data. Then an image stitching method was proposed to remove the overlapped data based on the normalised cross correlation (NCC) method. In order to detect the geometrical parameters of slub yarn, the frequency histogram , curve fitting, and spectrogram methods were adopted to analyse the sequence diameter data obtained. Four kinds of slub yarn with different geometrical parameters were tested using the method proposed and Uster method. The experimental results show that the detection results for slub amplitude, slub length, slub distance, and slub period obtained using the method proposed were consistent with the set values and Uster results.

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 segme...

Purpose The purpose of this paper is to propose a novel method using support vector machine (SVM) classifiers for objective seam pucker evaluation. Features are extracted using wavelet analysis and gray-level co-occurrence matrix (GLCM), and the samples are evaluated using SVM classifiers. The study aims to solve the problem of inappropriate parameters and large required samples in objective seam pucker evaluation. Design/methodology/approach Initially, seam pucker image was captured, and Edge detection and Hough transform were utilized to normalize the seam position and orientation. After cropping the image, the intensity was adjusted to the same identical level through histogram specification. Then, the standard deviations of the horizontal image and diagonal image, reconstructed using wavelet decomposition and reconstruction, were calculated based on parameter optimization. Meanwhile, GLCM was extracted from the restructured horizontal detail image, then the contrast and correlation of GLCM were calculated. Finally, these four features were imported to SVM classifiers based on genetic algorithm for evaluation. Findings The four extracted features reflected linear relationships among five grades. The experimental results showed that the classification accuracy was 96 percent, which catches up to the performance of human vision, and resolves ambiguity and subjective of the manual evaluation. Originality/value There are large required samples in current research. This paper provides a novel method using finite samples, and the parameters of the methods were discussed for parameter optimization. The evaluation results can provide references for analyzing the reason of wrinkles during garment manufacturing.

In the context of “double carbon”, as a traditional high energy consumption industry, the textile industry is facing the severe challenges of energy saving and emission reduction. To improve production efficiency in the textile industry, we propose the use of content-based image retrieval technology to shorten the fabric production cycle. However, fabric retrieval has high requirements for results, which makes it difficult for common retrieval methods to be directly applied to fabric retrieval. This paper presents a novel method for fabric image retrieval. Firstly, we define a fine-grained similarity to measure the similarity between two fabric images. Then, a convolutional neural network with a compact structure and cross-domain connections is designed to narrow the gap between fabric images and similarities. To overcome the problems of probabilistic missing and difficult training in classical hashing, we introduce a variational network module and structural module into the hashing model, which is called DVSH. We employ list-wise learning to perform similarity embedding. The experimental results demonstrate the superiority and efficiency of the proposed hashing model, DVSH.

The traditional manual defect detection method has low efficiency and is time-consuming and laborious. To address this issue, this paper proposed an automatic detection framework for fabric defect detection, which consists of a hardware system and detection algorithm. For the efficient and high-quality acquisition of fabric images, an image acquisition assembly equipped with three sets of lights sources, eight cameras, and a mirror was developed. The image acquisition speed of the developed device is up to 65 m per minute of fabric. This study treats the problem of fabric defect detection as an object detection task in machine vision. Considering the real-time and precision requirements of detection, we improved some components of CenterNet to achieve efficient fabric defect detection, including the introduction of deformable convolution to adapt to different defect shapes and the introduction of i-FPN to adapt to defects of different sizes. Ablation studies demonstrate the effectiveness of our proposed improvements. The comparative experimental results show that our method achieves a satisfactory balance of accuracy and speed, which demonstrate the superiority of the proposed method. The maximum detection speed of the developed system can reach 37.3 m per minute, which can meet the real-time requirements.

As a special case in content-based image retrieval, fabric retrieval has high potential application value in many fields. However, fabric retrieval has higher requirements for results, which makes it difficult for common retrieval methods to be directly applied to fabric retrieval. It is also a challenging issue with several obstacles: variety and complexity of fabric appearance, and high requirements for retrieval accuracy. To address this issue, this paper presents a novel method for fabric image retrieval based on soft similarity and pairwise learning. First, a soft similarity between two fabric images is defined to describe their relationship. Then, a convolutional neural network with compact structure and cross-domain connections is designed to learn the fabric image representation. Finally, listwise learning is introduced to train the convolutional neural network model and hash function. The generated hash codes are used to index the fabric image. The experiments are conducted on a wool fabric dataset. The experimental results show that the newly proposed method has a greater improvement than our previous work.

Factories must proof different colors on fabric in order to verify the feasibility of the color scheme. This proofing process is time-consuming and laborious. To recolor printed fabric images, a novel image recoloring method was proposed in this paper. This method can provide convenience for printed fabric designers and can be used to simulate printed fabric proofing. The relative total variation model was implemented to remove the fabric image texture and noise. Next, in the CIE1976 L* a* b* color space, the mean-shift clustering algorithm was utilized to segment the reference image and the target image to obtain the separated color regions. Then, the color regions of the reference image and the target image were matched based on the salient map and the values of the L* a* b* channels. Finally, local color transfer was performed between the matched color regions based on the matching results. Experiments were conducted on 100 printed fabrics with different color schemes. Results indicated that the proposed method can transfer the color appearance of the reference printed fabric image to the target printed fabric image and realize printed fabric image recoloring.

Due to the potential values in many areas such as e-commerce and inventory management, fabric image retrieval, which is a special case in Content Based Image Retrieval (CBIR), has recently become a research hotspot. It is also a challenging issue with serval obstacles: variety and complexity of fabric appearance, high requirements for retrieval accuracy. To address this issue, this paper proposes a novel approach for fabric image retrieval based on multi-task learning and deep hashing. According to the cognitive system of fabric, a multi-classification-task learning model with uncertainty loss and constraint is presented to learn fabric image representation. Then we adopt an unsupervised deep network to encode the extracted features into 128-bits hashing codes. Further, the hashing codes are regarded as the index of fabrics image for image retrieval. To evaluate the proposed approach, we expanded and upgraded the dataset WFID, which was built in our previous research specifically for fabric image retrieval. The experimental results show that the proposed approach outperforms the state-of-the-art.