A contrastive learning-based attention generative adversarial network for defect detection in colour-patterned fabric in COLORATION TECHNOLOGY, Vol. 139, N° 3 (06/2023)  

[article]  inCOLORATION TECHNOLOGY > Vol. 139, N° 3 (06/2023) . - p. 248-264 Titre : A contrastive learning-based attention generative adversarial network for defect detection in colour-patterned fabric Type de document : texte imprimé Année de publication : 2023 Article en page(s) : p. 248-264 Note générale : Bibliogr. Langues : Anglais (eng) Catégories : Apprentissage contrastif Détection de défauts (Ingénierie) Textiles et tissus -- Défauts Textiles et tissus teints Index. décimale : 667.3 Teinture et impression des tissus Résumé : The pattern style of colour-patterned fabrics is varied. Defective fabric samples are scarce in the production of small batches of colour-patterned fabrics. Therefore, the unsupervised defect-detection method for colour-patterned fabric has attracted wide attention. Several unsupervised defect-detection methods for colour-patterned fabrics based on convolutional neural networks have been proposed. However, convolutional neural network methods cannot learn long-range semantic information interaction well because of the intrinsic locality of convolution operations. Besides, as the number of layers in the convolutional neural network increases, the feature maps become more and more complex. Convolutional neural networks experience difficulties in coordinating numerous parameters and extracting key features from complex feature maps. Both these problems reduce the accuracy of the model for detecting defects in colour-patterned fabrics. In this paper, we propose a Contrastive Learning-based Attention Generative Adversarial Network (CLAGAN) for defect detection in colour-patterned fabrics. The CLAGAN possesses two important parts: contrastive learning and a channel attention module. Contrastive learning captures long-range dependencies by calculating the cosine similarity between different features. The channel attention module assigns different weights to each channel of the feature maps, and it enables the model to extract key features from those feature maps. The experimental results verified the effectiveness of the CLAGAN. It obtained values of 38.25% for intersection over union and of 51.67% for the F1-measure on the YDFID-2 public dataset. Note de contenu : - RELATED WORKS : Fabric defect-detection methods based on deep learning - Contrastive learning - Attention mechanism - THE PROPOSED CLAGAN MODEL : An overview of the CLAGAN - Defect detection in colour-patterned fabric based on the CLAGAN - RESULTS AND DISCUSSION : Datasets - Evaluation metric - Implementation detail - Experimental results and analysis - Table 1 : The number of colour-patterned fabric samples - Table 2 : Quantitative detection results for different methods on the YDFID-1 dataset - Table 3 : Quantitative detection results for different methods on the ZJU-Leaper dataset - Table 4 : Experimental results for the noise fraction - Table 5 : Experimental results for the ablation study DOI : https://doi.org/10.1111/cote.12642 En ligne : https://onlinelibrary.wiley.com/doi/epdf/10.1111/cote.12642 Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=39536 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24086	-	Périodique	Bibliothèque principale	Documentaires	Disponible

Evolving regularised random vector functional link by seagull optimisation algorithm for yarn-dyed fabric colour difference classification / Yufeng Qiu in COLORATION TECHNOLOGY, Vol. 140, N° 3 (06/2024)  

[article]  inCOLORATION TECHNOLOGY > Vol. 140, N° 3 (06/2024) . - p. 467-482 Titre : Evolving regularised random vector functional link by seagull optimisation algorithm for yarn-dyed fabric colour difference classification Type de document : texte imprimé Auteurs : Yufeng Qiu, Auteur ; Zhiyu Zhou, Auteur ; Jianxin Zhang, Auteur Année de publication : 2024 Article en page(s) : p. 467-482 Note générale : Bibliogr. Langues : Anglais (eng) Catégories : Algorithme d'optimisation Les algorithmes d’optimisation cherchent à déterminer le jeu de paramètres d’entrée d’une fonction donnant à cette fonction la valeur maximale ou minimale. On cherchera par exemple la découpe optimale d’une tôle pour en fabriquer le plus grand nombre de boîtes de conserve possible (ou d’un tissu pour en faire le plus grand nombre de chemises possibles, etc.). Cette optimisation peut se faire sans contrainte ou sous contrainte, le second cas se ramenant au premier dans le cas des fonctions dérivables par la méthode du multiplicateur de Lagrange (et des fonctions non-dérivables par l’algorithme d’Everett). - Algorithmes : Le problème est insoluble en tant que tel si l’on ne connaît rien de la fonction (il existe peut-être une combinaison très particulière de valeurs d’entrées lui donnant ponctuellement une valeur extrêmement haute ou basse, qui pourrait échapper à l’algorithme. Aussi existe-t-il plusieurs classes d’algorithmes liés aux différentes connaissances qu’on peut avoir sur la fonction. Si celle-ci est dérivable, l’une des plus performantes est celle du gradient conjugué. Aucune méthode connue en 2004 (à part l’énumération exhaustive ou l’analyse algébrique) ne permet de trouver avec certitude un extremum global d’une fonction. Les extrema déterminables sont toujours locaux à un domaine, et demandent souvent même en ce cas quelques caractéristiques à la fonction, par exemple dans certains cas la continuité. Les métaheuristiques sont une classe d’algorithmes d’optimisation qui tentent d’obtenir une valeur approchée de l’optimum global dans le cas de problèmes d’optimisation difficile. Elles ne donnent cependant aucune garantie sur la fiabilité du résultat. (Wikipedia) Colorimétrie Couleurs -- Classification Textiles et tissus teints Index. décimale : 535.6 Couleur Résumé : To address the issue of low precision in classifying the colour differences of yarn-dyed fabrics and the high cost of manual detection, a colour difference classification method relying on an improved seagull optimisation algorithm (SOA) optimised regularised random vector functional link (RRVFL) model is proposed for dyed fabrics. First, to address the issue of the slow convergence speed of the SOA, the current study optimises the initial SOA group with the marine predators algorithm (MPA) so that it can effectively improve the convergence ability and global optimisation ability of the SOA. Subsequently, the enhanced SOA is applied to fine-tune the parameters of the RRVFL. Compared with the methods that only optimise weights and bias, the proposed algorithm obtained by optimizing the initial group of SOA through the Marine Predators Algorithm (MSOA)-RRVFL model in this paper also increases the optimisation of the number of nodes in the hidden layer and regularisation parameters, which also effectively avoids the issue of the low classification accuracy of the RRVFL model due to random related parameters. Finally, by comparing the RRVFL model with other optimisation algorithms, the experimental outcomes demonstrate that the convergence ability of the improved SOA has been improved, and that the average accuracy of colour difference classification by the MSOA-RRVFL model is as high as 99.79%, and that the classification error fluctuation can be stabilised below 0.2%. In general, the MSOA-RRVFL model displays an excellent performance in terms of stability and significance. Note de contenu : - YARN-DYED FABRIC COLOUR DIFFERENCE CLASSIFICATION WITH RRVFL OPTIMISED BY MSOA : Seagull optimisation algorithm - Marine predators algorithm - RRVFL network - Classifying the dyed fabric colour difference with the proposed MSOA-RRVFL - EXPERIMENTAL RESULTS AND ANALYSIS : Acquisition of the experimental data - Parameter selection for the SOA - Selection of activation function, number of hidden ganglia and regularisation factor of the RRVFL - Stability analysis of the algorithms - Significance analysis - Computing complexity DOI : https://doi.org/10.1111/cote.12722 En ligne : https://drive.google.com/file/d/1t0YQpUySjS2GsCkVUGJV328RF4Z8Imvx/view?usp=drive [...] Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=40963 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24657	-	Périodique	Bibliothèque principale	Documentaires	Disponible

Instrumental shade sorting of coloured fabrics using genetic algorithm and particle swarm optimisation / Elham Hasanlou in COLORATION TECHNOLOGY, Vol. 139, N° 4 (08/2023)  

[article]  inCOLORATION TECHNOLOGY > Vol. 139, N° 4 (08/2023) . - p. 454-463 Titre : Instrumental shade sorting of coloured fabrics using genetic algorithm and particle swarm optimisation Type de document : texte imprimé Auteurs : Elham Hasanlou, Auteur ; Ali Shams Nateri, Auteur ; Hossein Izadan, Auteur Année de publication : 2023 Article en page(s) : p. 454-463 Note générale : Bibliogr. Langues : Anglais (eng) Catégories : Algorithmes génétiques Colorimétrie Dispositifs de tri Optimisation par essaims particulaires L'optimisation par essaims particulaires (OEP ou PSO en anglais) est une métaheuristique d'optimisation, inventée par Russel Eberhart (ingénieur en électricité) et James Kennedy (socio-psychologue) en 1995. Algorithme Cet algorithme s'inspire à l'origine du monde du vivant. Il s'appuie notamment sur un modèle développé par Craig Reynolds à la fin des années 1980, permettant de simuler le déplacement d'un groupe d'oiseaux. Une autre source d'inspiration, revendiquée par les auteurs, James Kennedy et Russel Eberhart, est la socio-psychologie. Cette méthode d'optimisation se base sur la collaboration des individus entre eux. Elle a d'ailleurs des similarités avec les algorithmes de colonies de fourmis, qui s'appuient eux aussi sur le concept d'auto-organisation. Cette idée veut qu'un groupe d'individus peu intelligents peut posséder une organisation globale complexe. Ainsi, grâce à des règles de déplacement très simples (dans l'espace des solutions), les particules peuvent converger progressivement vers un minimum global. Cette métaheuristique semble cependant mieux fonctionner pour des espaces en variables continues. (Wikipedia) Textiles et tissus teints Index. décimale : 667.3 Teinture et impression des tissus Résumé : In the present research by combination of Clemson Colour Clustering (CCC) instrumental shade sorting method and two metaheuristic algorithms, a genetic algorithm (GA) and a particle swarm optimisation (PSO), two new shade sorting methods, called CCCGA and CCCPSO were proposed. Then these proposed methods were applied on 16 well-prepered colour sets made of coloured fabrics and their results were compared using some important performance evaluation factors. The results of the methods were also compared with conventional CCC shade sorting method and a method based on CCC combined with k-means technique (CCCk). The results obtained from various shade sorting methods showed that the CCCGA and CCCPSO methods successfully sorted the coloured fabrics with high efficiency, and their results slightly outperformed the results of the CCC method. Note de contenu : - INTRODUCTION : Genetic algorithm - Particle swarm optimisation - MATERIALS AND METHODS : Preparation of samples and colour measurement - Determining the colour tolerance of coloured fabrics - Shade sorting using the genetic algorithm - Shade sorting using particle swarm optimisation - RESULTS AND DISCUSSION : Performance evaluation factors of shade sorting methods - Number of sorted groups - Colour variation within the groups - Compactness of the points in sorted groups - Utilisation of the fabric - Table 1 : Specifications and colour tolerance of the 16 fabric colour sets - Table 2 : The values of the parameters and genetic operators used in the genetic algorithm - Table 3 : The values of the parameters used in the particle swarm optimisation - Table 4 : The colour variation within the groups formed by various shade sorting methods and the number of sorted groups by Clemson Colour Clustering (CCC) shade sorting - Table 5 : The compactness of the points in sorted groups by different shade sorting methods - Table 6 : The number of groups containing only one sample and the percentage of groups in low utilisation by different shade sorting methods DOI : https://doi.org/10.1111/cote.12663 En ligne : https://onlinelibrary.wiley.com/doi/epdf/10.1111/cote.12663 Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=39682 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24152	-	Périodique	Bibliothèque principale	Documentaires	Disponible

Knowledge distillation for unsupervised defect detection of yarn-dyed fabric using the system DAERD : Dual attention embedded reconstruction distillation / Hongwei Zhang in COLORATION TECHNOLOGY, Vol. 140, N° 1 (02/2024)  

[article]  inCOLORATION TECHNOLOGY > Vol. 140, N° 1 (02/2024) . - p. 125-143 Titre : Knowledge distillation for unsupervised defect detection of yarn-dyed fabric using the system DAERD : Dual attention embedded reconstruction distillation Type de document : texte imprimé Auteurs : Hongwei Zhang, Auteur ; Shuaibo Liu, Auteur ; Shuai Lu, Auteur ; Le Yao, Auteur ; Pengfei Li, Auteur Année de publication : 2024 Article en page(s) : p. 125-143 Note générale : Bibliogr. Langues : Anglais (eng) Catégories : Détection de défauts (Ingénierie) Textiles et tissus teints Index. décimale : 667.3 Teinture et impression des tissus Résumé : Detecting defects of yarn-dyed fabrics automatically in industrial scenarios can improve economic efficiency, but the scarcity of defect samples makes the task more challenging in the customised and small-batch production scenario. At present, most reconstruction-based methods have high requirements on the effect of reconstructing the defect area into the normal area, and the reconstruction performance often determines the final defect detection result. To solve this problem, this article proposes an unsupervised learning framework of dual attention embedded reconstruction distillation. We try to use this novel distillation scheme to provide some contribution to the defect detection field. Firstly, different from the encoder-encoder structure of traditional distillation, the teacher-student network in this article adopts the encoder-decoder structure. The purpose of the student network is to restore the normal feature representation of the pre-trained teacher network. Secondly, this article proposes a dual attention residual module, which can effectively remove redundant information and defective feature information from the teacher network through the double feature weight allocation mechanism. This helps the student network to recover the normal feature information output by the teacher network. Finally, the multi-level training deployment at the feature level in this article aims to make the model obtain accurate defect detection results. The proposed method has been extensively tested on the published fabric dataset YDFID-1, ZJU-Leaper dataset and the anomaly detection dataset MVTec. The results show that this method not only has good performance in fabric defect detection and location but also has universal applicability. Note de contenu : - RELATED WORKS : Fabric defect detection - Attention mechanism - METHODS : The proposed DAERD model - DAERD-based yarn-dyed fabric defect detection - EXPERIMENT : Experimental details - Datasets - Evaluation metric - Experimental results and comparative analysis - Table 1 : Distribution of the number of defective and defect-free samples in the partial dataset of YDFID-1 - Table 2 : Distribution of the number of anomaly-free and anomalous samples in the texture class dataset - Table 3 : Image-level AUROC (%) of defect detection methods on the YDFID-1 dataset - Table 4 : Pixel-level AUROC (%) and AUPRO (%) of defect localisation methods on the YDFID-1 dataset - Table 5 : Image-level AUROC (%) of defect detection methods on the MVTec dataset - Table 6 : Pixel-level AUROC (%) and AUPRO (%) of defect localisation methods on the MVTec dataset - Table 7 : Image-level AUROC (%) of defect detection methods on the ZJU-Leaper dataset - Table 8 : Pixel-level AUROC (%) and AUPRO (%) of defect localisation methods on the ZJU-Leaper dataset - Table 9 : Ablation study of the effect of SimAM (simple, parameter-free, attention module) and SE (squeeze-and-excitation) network on the model performance DOI : https://doi.org/10.1111/cote.12705 En ligne : https://drive.google.com/file/d/1fzssAGVLxuRnKx8FtikkhnIF1yv1cue3/view?usp=drive [...] Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=40396 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24413	-	Périodique	Bibliothèque principale	Documentaires	Disponible

Mechanism and application of ozone fading: Oxidative decolorisation of disperse dyes and waste-dyed polyester fabrics / Jiangfei Lou in COLORATION TECHNOLOGY, Vol. 139, N° 3 (06/2023)  

[article]  inCOLORATION TECHNOLOGY > Vol. 139, N° 3 (06/2023) . - p. 338-349 Titre : Mechanism and application of ozone fading: Oxidative decolorisation of disperse dyes and waste-dyed polyester fabrics Type de document : texte imprimé Auteurs : Jiangfei Lou, Auteur ; Lamei Ren, Auteur ; Jiugang Yuan, Auteur ; Jing Xu, Auteur ; Zhengbiao Gu, Auteur ; Xuerong Fan, Auteur Année de publication : 2023 Article en page(s) : p. 338-349 Note générale : Bibliogr. Langues : Anglais (eng) Catégories : Bleu (couleur) Caractérisation Colorants -- Oxydation Décoloration Fibres textiles -- Surfaces Mesure Orange (couleur) Ozone Polyesters Textiles et tissus teints Index. décimale : 667.3 Teinture et impression des tissus Résumé : As the most common synthetic fibre, polyester has consistently been in high demand and usage, but its recycling has always been a problem for the textile industry. The decolorisation of dyed polyester is the key to expanding its recycling use, and there is no rapid, effective and simple decolorisation method. In this study, ozone was used to decolorise dispersed dyes developed for polyester. Ozone was effective in decolorising CI Disperse Orange 30 and CI Disperse Blue 60, and the process was extended to the decolorisation of dyed polyester fabric. Based on single factor analysis, the optimum decolorisation technological conditions were: an ozone feed rate of 130 mg L−1 min−1, treatment time of 2 hours, wet pick-up of 100% and pH value of 3, with which the decolorisation rate and colour difference values could exceed 45% and 6%, respectively. The changes in dye structure during the process of ozone decolorisation were analysed using scanning electron microscopy, X-ray diffraction and Fourier Transform–infrared spectroscopy, and the mechanism of ozone decolorisation was also investigated. The results showed that the crystallinity decreased from 41.18% to 36.30% with a strength retention rate of above 80%, and demonstrated that ozone can be used on polyester fabric and dye simultaneously. The aim of entering the fibre was achieved by etching the polyester and decolorisation was achieved by oxidising the chromaticity groups. Note de contenu : - EXPERIMENTAL : Materials and chemicals - Dyeing of the PET fabrics - Monitoring of the ozone feed rate - Optimum decolorisation technological conditions - Exploration of the decolorisation mechanism - Characterisation and measurement - RESULTS AND DISCUSSION : Optimum decolorisation technological conditions - Application of the optimum decolorisation technological conditions - Surface morphology of PET fabrics treated with ozone - Decolorisation of dispersed dye with ozone - Table 1 : Names, classes and structures of the dyes used in this study - Table 2 : Ozone feed rates used in this study - Table 3 : Dyeing conditions of treated samples - Table 4 : CIElab values of the ozone-treated and reference-dyed PET DOI : https://doi.org/10.1111/cote.12654 En ligne : https://onlinelibrary.wiley.com/doi/epdf/10.1111/cote.12654 Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=39543 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24086	-	Périodique	Bibliothèque principale	Documentaires	Disponible

---

1 (1 - 5 / 5)