Pub Date : 2023-02-28DOI: 10.35530/it.074.01.202224
Peng Tao, Cao Wenli, Chen Jia, LV Xinghang, Zhang Zili, Liu Junping, Hu Xinrong
Fabric classification plays a crucial role in the modern textile industry and fashion market. In the early stage, traditional neural network methods were adopted to identify fabrics with the drawback of restricted fabric type and poor accuracy. Combining multi-frame temporality and analysing fabric graph data made from fabric motion features, this paper proposes a novel hybrid model that introduces the concept of graph networks to classify 30 textile materials in a public database. We utilize the graph inductive representation learning method (GraphSAGE, Graph Sample and Aggregate) to extract node embedding features of the fabric. Moreover, bidirectional gated recurrent unit and layer attention mechanism (BiGRU-attention) are employed in the last layer of aggregation to calculate the score of previous cells. Intending to further enhance performance, we link the jump connection with adaptive selection aggregation frameworks to determine the influential region of each node. Our method breaks through the limitation that the original methods can only classify a few fabrics with great classification results.
{"title":"Research on fabric classification based on graph neural network","authors":"Peng Tao, Cao Wenli, Chen Jia, LV Xinghang, Zhang Zili, Liu Junping, Hu Xinrong","doi":"10.35530/it.074.01.202224","DOIUrl":"https://doi.org/10.35530/it.074.01.202224","url":null,"abstract":"Fabric classification plays a crucial role in the modern textile industry and fashion market. In the early stage, traditional\u0000neural network methods were adopted to identify fabrics with the drawback of restricted fabric type and poor accuracy.\u0000Combining multi-frame temporality and analysing fabric graph data made from fabric motion features, this paper\u0000proposes a novel hybrid model that introduces the concept of graph networks to classify 30 textile materials in a public\u0000database. We utilize the graph inductive representation learning method (GraphSAGE, Graph Sample and Aggregate)\u0000to extract node embedding features of the fabric. Moreover, bidirectional gated recurrent unit and layer attention\u0000mechanism (BiGRU-attention) are employed in the last layer of aggregation to calculate the score of previous cells.\u0000Intending to further enhance performance, we link the jump connection with adaptive selection aggregation frameworks\u0000to determine the influential region of each node. Our method breaks through the limitation that the original methods can\u0000only classify a few fabrics with great classification results.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49002128","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.202287
M. Špiler, D. Milosevic, Miroslav Miskic, Ladin Gostimirović, Milan Beslać, Boris Jevtić
The investments in digital technologies are expected to soon have a major impact on the textile and fashion companies’ sustainability and competitiveness. Motivated by these trends empirical research on investments of the fashion and textile companies in ICT technologies-based advancement in the Serbian case was provided in 2022. Representatives of 423 textile and fashion companies were asked about their investments in various digital technologies in the previous three years and their digital transformation status. The research findings show that investments in cloud computing, IT, energy management, automation, robotics, and machine learning technologies have a significant impact on the digital transformation of companies. Most of them reached a medium level of transformation, fewer than a high level, with many textile and fashion companies just defining digital transformation. The contribution of the research findings to the investments in the companies’ digital transformation can be seen in the significance of the textile’s digital technology implementation, which enables manufacturers and retailers to respond directly to market demand by reducing product lead time and cost, increasing supply chain efficiency and profitability, and promising in terms of ensuring competitive advantage in the risk and challenging business environment.
{"title":"Investments in digital technology advances in textiles","authors":"M. Špiler, D. Milosevic, Miroslav Miskic, Ladin Gostimirović, Milan Beslać, Boris Jevtić","doi":"10.35530/it.074.01.202287","DOIUrl":"https://doi.org/10.35530/it.074.01.202287","url":null,"abstract":"The investments in digital technologies are expected to soon have a major impact on the textile and fashion companies’\u0000sustainability and competitiveness. Motivated by these trends empirical research on investments of the fashion and\u0000textile companies in ICT technologies-based advancement in the Serbian case was provided in 2022. Representatives\u0000of 423 textile and fashion companies were asked about their investments in various digital technologies in the previous\u0000three years and their digital transformation status. The research findings show that investments in cloud computing, IT,\u0000energy management, automation, robotics, and machine learning technologies have a significant impact on the digital\u0000transformation of companies. Most of them reached a medium level of transformation, fewer than a high level, with many\u0000textile and fashion companies just defining digital transformation. The contribution of the research findings to the\u0000investments in the companies’ digital transformation can be seen in the significance of the textile’s digital technology\u0000implementation, which enables manufacturers and retailers to respond directly to market demand by reducing product\u0000lead time and cost, increasing supply chain efficiency and profitability, and promising in terms of ensuring competitive\u0000advantage in the risk and challenging business environment.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47158475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Toma, G. Popescu, A. Popescu, S. Olaru, A. Săliștean, I. Badea
Emergency workers are exposed to many different risks at the same time and possible consequences for their safety and health may be manifold. Many emergency workers suffer from accidents and injuries in the course of their jobs, as well as other negative health effects that lead to severe deterioration of their physical and psychological well-being. The use of specific personal protective equipment (PPE) according to the given risks is of great importance in preventing adverse health effects among emergency workers. This research aimed to develop, for emergency workers, a PPE system, in a modular structure consisting of: i) modular layer 1: the inner layer, in contact with the skin/Underwear PPE, with the function of sensorial and thermophysiological comfort and which ensures thermal protection; ii) modular layer 2: the intermediate (basic) layer/Duty uniform – with the function of limited protection to the specific risk factors of an unpredictable intervention action (thermal risks: convection heat, flame; risks from the external environment: liquid splashes; mechanical risks: cutting, abrasion, etc); iii) modular layer 3: the outer layer/specialized PPE, with a function of barrier against specific risk factors for fire intervention missions, extreme weather conditions etc. This modular approach provides some advantages, including preserving comfort and flexibility until the intervention mission requires the use of the next level of protection. This helps ensure that emergency responders are not in the position of choosing between their safety or mission effectiveness.
{"title":"Protective clothing system for interventions in emergency situations","authors":"D. Toma, G. Popescu, A. Popescu, S. Olaru, A. Săliștean, I. Badea","doi":"10.35530/it.074.01.1839","DOIUrl":"https://doi.org/10.35530/it.074.01.1839","url":null,"abstract":"Emergency workers are exposed to many different risks at the same time and possible consequences for their safety\u0000and health may be manifold. Many emergency workers suffer from accidents and injuries in the course of their jobs, as\u0000well as other negative health effects that lead to severe deterioration of their physical and psychological well-being. The\u0000use of specific personal protective equipment (PPE) according to the given risks is of great importance in preventing\u0000adverse health effects among emergency workers. This research aimed to develop, for emergency workers, a PPE\u0000system, in a modular structure consisting of: i) modular layer 1: the inner layer, in contact with the skin/Underwear PPE,\u0000with the function of sensorial and thermophysiological comfort and which ensures thermal protection; ii) modular layer 2:\u0000the intermediate (basic) layer/Duty uniform – with the function of limited protection to the specific risk factors of an\u0000unpredictable intervention action (thermal risks: convection heat, flame; risks from the external environment: liquid\u0000splashes; mechanical risks: cutting, abrasion, etc); iii) modular layer 3: the outer layer/specialized PPE, with a function\u0000of barrier against specific risk factors for fire intervention missions, extreme weather conditions etc. This modular\u0000approach provides some advantages, including preserving comfort and flexibility until the intervention mission requires\u0000the use of the next level of protection. This helps ensure that emergency responders are not in the position of choosing\u0000between their safety or mission effectiveness.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41493385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.202252
Han Chen, Han Xu, Yudian Zhang, Weifan Wang, Zhengyang Lu
Garment heritages are commonly missing evidence of restoration because of their age and complex preservation environment. Traditional restoration methods rely on the subjective experience of restoration personnel. Its restoration results are difficult to achieve accuracy and objectivity, exposing precious cultural relics to the risk of irreversible secondary damage. Taking the Pale Brown Lace-encrusted Unlined Coat as an example, this study puts forward a method of garment heritages restoration based on digital virtual technology. After fully researching the garment background information, we used deep learning and virtual twin technology to draw and cut the garment pieces, arrange and sew the garments, and complete the stained patterns. The results show that our method can restore the original appearance of the heritages relatively well, providing a new method reference for the inheritance and digital transmission of traditional garment heritages.
{"title":"The restoration of garment heritages based on digital virtual technology:\u0000A case of the Chinese pale brown lace-encrusted unlined coat","authors":"Han Chen, Han Xu, Yudian Zhang, Weifan Wang, Zhengyang Lu","doi":"10.35530/it.074.01.202252","DOIUrl":"https://doi.org/10.35530/it.074.01.202252","url":null,"abstract":"Garment heritages are commonly missing evidence of restoration because of their age and complex preservation\u0000environment. Traditional restoration methods rely on the subjective experience of restoration personnel. Its restoration\u0000results are difficult to achieve accuracy and objectivity, exposing precious cultural relics to the risk of irreversible\u0000secondary damage. Taking the Pale Brown Lace-encrusted Unlined Coat as an example, this study puts forward a\u0000method of garment heritages restoration based on digital virtual technology. After fully researching the garment\u0000background information, we used deep learning and virtual twin technology to draw and cut the garment pieces, arrange\u0000and sew the garments, and complete the stained patterns. The results show that our method can restore the original\u0000appearance of the heritages relatively well, providing a new method reference for the inheritance and digital\u0000transmission of traditional garment heritages.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43320676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.202268
Baoru Ge, N. Shaari, Mohd Yazid Mohd Yunos, S. Abidin
The sales growth rate of men's plain-colour shirts dropped significantly online in China. Consumers first pay attention to the appearance design of clothing online. It only takes 7 seconds for consumers to determine a product, and the colour in its appearance design accounts for about 67% of the role. Thus, this study took the colour design of men's plain-colour shirts as an example in China, established the basic colour calculation scale and an algorithm model of group consumers' product preferences based on Kansei Engineering and scientific mathematics, to provide new sales ideas and methods for retailers and markets online. Firstly, this study obtained the crucial Kansei word pairs (emotional preferences) and colour design elements through interviews, literature, magazines and websites, word frequency statistics, card sorting and cluster analysis. Then, researchers established a basic colour calculation scale of cross-loading through Kansei Engineering and partial least squares (PLS). Finally, a recommendation set of products is obtained using the analytic hierarchy process (AHP), the weight of Kansei word pairs, and the distance calculation of comprehensive evaluation value based on consumers' emotional needs. That is, this study obtained consumers' aesthetic emotional preference for men's plain-colour shirts in China, colour design elements of shirts that are widely recognized and accepted, basic colour calculation scales, recommendation preferences algorithms and models for group consumers, and verified their effectiveness by PCA.
{"title":"Group consumers' preference recommendation algorithm model for online\u0000apparel's colour based on Kansei engineering","authors":"Baoru Ge, N. Shaari, Mohd Yazid Mohd Yunos, S. Abidin","doi":"10.35530/it.074.01.202268","DOIUrl":"https://doi.org/10.35530/it.074.01.202268","url":null,"abstract":"The sales growth rate of men's plain-colour shirts dropped significantly online in China. Consumers first pay attention to\u0000the appearance design of clothing online. It only takes 7 seconds for consumers to determine a product, and the colour\u0000in its appearance design accounts for about 67% of the role. Thus, this study took the colour design of men's plain-colour\u0000shirts as an example in China, established the basic colour calculation scale and an algorithm model of group\u0000consumers' product preferences based on Kansei Engineering and scientific mathematics, to provide new sales ideas\u0000and methods for retailers and markets online. Firstly, this study obtained the crucial Kansei word pairs (emotional\u0000preferences) and colour design elements through interviews, literature, magazines and websites, word frequency\u0000statistics, card sorting and cluster analysis. Then, researchers established a basic colour calculation scale of\u0000cross-loading through Kansei Engineering and partial least squares (PLS). Finally, a recommendation set of products is\u0000obtained using the analytic hierarchy process (AHP), the weight of Kansei word pairs, and the distance calculation of\u0000comprehensive evaluation value based on consumers' emotional needs. That is, this study obtained consumers'\u0000aesthetic emotional preference for men's plain-colour shirts in China, colour design elements of shirts that are widely\u0000recognized and accepted, basic colour calculation scales, recommendation preferences algorithms and models for\u0000group consumers, and verified their effectiveness by PCA.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47125411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.2021111
Yanbo Ji, Ying Wang, Kaixuan Liu, Mengyue Hu, Chun Zhu, Zhao Lü, Xiaoning Li
Based on the human torso point cloud, this paper proposes a method from the 3D design of the corset to the 2D pattern expansion. The point cloud of the human body is obtained through 3D scanning. The human body model for research is constructed, and the 3D basic style design of the corset is carried out, based on the same style and different structural line design, and through the curved surface flattening platform to convert 3D into 2D patterns. The verification was made through a virtual simulation platform and physical production methods. This study enriches the application prospect of digital technology in clothing design. Our proposed solution provides a more intuitive wedding dress design method and improves fit and comfort. It can significantly reduce the difficulty of wedding pattern-making and improve the efficiency of wedding design. In addition, our proposed method is not only suitable for wedding dress design, but also other styles of clothing design.
{"title":"3D interactive design of wedding dress","authors":"Yanbo Ji, Ying Wang, Kaixuan Liu, Mengyue Hu, Chun Zhu, Zhao Lü, Xiaoning Li","doi":"10.35530/it.074.01.2021111","DOIUrl":"https://doi.org/10.35530/it.074.01.2021111","url":null,"abstract":"Based on the human torso point cloud, this paper proposes a method from the 3D design of the corset to the 2D pattern\u0000expansion. The point cloud of the human body is obtained through 3D scanning. The human body model for research\u0000is constructed, and the 3D basic style design of the corset is carried out, based on the same style and different structural\u0000line design, and through the curved surface flattening platform to convert 3D into 2D patterns. The verification was made\u0000through a virtual simulation platform and physical production methods. This study enriches the application prospect of\u0000digital technology in clothing design. Our proposed solution provides a more intuitive wedding dress design method and\u0000improves fit and comfort. It can significantly reduce the difficulty of wedding pattern-making and improve the efficiency\u0000of wedding design. In addition, our proposed method is not only suitable for wedding dress design, but also other styles\u0000of clothing design.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46837765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.2022148
Malina Rosca, ANA-DIANA Vatra, M. Avadanei
Many clothing companies approach digital transformation by focusing on digitizing individual processes or operations. Digital transformation is often limited to specific initiatives or programmes that only impact a few departments. Significant opportunities or existential risks are often the main drivers for digital transformation. Moreover, leaders planning the future of their companies and industries should focus on the opportunity – or existential threat – that these changes present. It is essential to find the ideal balance between focusing on quick results with innovative ideas and laying the foundation for digital transformation, such as unleashing the potential of data and analytics, managing brand and reputational risk, controlling the entire supply chain and closing the digital technology gaps are not the only significant issues. A complete change in corporate culture that puts the customer at the centre is the key component of the ultimate digital challenge for clothing companies. This article presents the opportunities, benefits and challenges of developing garment models with digital tools from Gemini CAD, a Lectra company. These tools include (in addition to the pattern) the product data sheet, a detailed description of all fabrics, trimmings, and accessories, components needed for sourcing, purchasing, and determining the cost of the product, as well as the information needed to publish the product on e-commerce and interact with the customer, including customization.
{"title":"The digital transformation of garment product development","authors":"Malina Rosca, ANA-DIANA Vatra, M. Avadanei","doi":"10.35530/it.074.01.2022148","DOIUrl":"https://doi.org/10.35530/it.074.01.2022148","url":null,"abstract":"Many clothing companies approach digital transformation by focusing on digitizing individual processes or operations.\u0000Digital transformation is often limited to specific initiatives or programmes that only impact a few departments. Significant\u0000opportunities or existential risks are often the main drivers for digital transformation. Moreover, leaders planning the\u0000future of their companies and industries should focus on the opportunity – or existential threat – that these changes\u0000present. It is essential to find the ideal balance between focusing on quick results with innovative ideas and laying the\u0000foundation for digital transformation, such as unleashing the potential of data and analytics, managing brand and\u0000reputational risk, controlling the entire supply chain and closing the digital technology gaps are not the only significant\u0000issues. A complete change in corporate culture that puts the customer at the centre is the key component of the ultimate\u0000digital challenge for clothing companies. This article presents the opportunities, benefits and challenges of developing\u0000garment models with digital tools from Gemini CAD, a Lectra company. These tools include (in addition to the pattern)\u0000the product data sheet, a detailed description of all fabrics, trimmings, and accessories, components needed for\u0000sourcing, purchasing, and determining the cost of the product, as well as the information needed to publish the product\u0000on e-commerce and interact with the customer, including customization.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43091513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.202279
Yuzhuo Li, Lei Jiang, Xinrong Li, W. Feng
Developing the technology of estimating human body size from two-dimensional images is the key to realising more digitalization and artificial intelligence in the textile and garment industry. Therefore, this paper is an in-depth study of estimating body sizes from two-dimensional images in a self-collected database of human body samples. First, the artificial thresholds in the Canny edge operator were replaced by adaptive thresholds. The improved Canny edge operator was combined with mathematical morphology so that it could detect a clear and complete single human contour. Then a joint point detection algorithm based on a convolution neural network and human proportion is proposed. It can detect human feature points with different body proportions. Finally, front and side images and manual body measurements of 122 males aged 18–22 years were collected as the human sample database, calculating the length and fit of the girth size. Compared with manual body measurement data, the error of human length and girth size parameters within the national standard range of –1.5 ~ 1.5 cm can reach 91% on average. This study provides an accurate and convenient anthropometric method for digital garment engineering, which can be used for online shopping and garment customization, and has a certain practical value.
{"title":"Non-contact clothing anthropometry based on two-dimensional image\u0000contour detection and feature point recognition","authors":"Yuzhuo Li, Lei Jiang, Xinrong Li, W. Feng","doi":"10.35530/it.074.01.202279","DOIUrl":"https://doi.org/10.35530/it.074.01.202279","url":null,"abstract":"Developing the technology of estimating human body size from two-dimensional images is the key to realising more\u0000digitalization and artificial intelligence in the textile and garment industry. Therefore, this paper is an in-depth study of\u0000estimating body sizes from two-dimensional images in a self-collected database of human body samples. First, the\u0000artificial thresholds in the Canny edge operator were replaced by adaptive thresholds. The improved Canny edge\u0000operator was combined with mathematical morphology so that it could detect a clear and complete single human\u0000contour. Then a joint point detection algorithm based on a convolution neural network and human proportion is\u0000proposed. It can detect human feature points with different body proportions. Finally, front and side images and manual\u0000body measurements of 122 males aged 18–22 years were collected as the human sample database, calculating the\u0000length and fit of the girth size. Compared with manual body measurement data, the error of human length and girth size\u0000parameters within the national standard range of –1.5 ~ 1.5 cm can reach 91% on average. This study provides an\u0000accurate and convenient anthropometric method for digital garment engineering, which can be used for online shopping\u0000and garment customization, and has a certain practical value.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43806613","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.202185
JIA-ZHEN Chen, ZI-YI Guo, Tao Li, Lei Du, F. Zou
The purpose of this study is to propose a new method to achieve pattern generation from garment sample without damage. The non-contact three-dimensional (3D) scanner was employed to get the point cloud data of garment samples. The Bowyer-Watson algorithm was used to implement Delaunay triangulation for surface reconstruction. The finite element (FE) approach was employed to achieve the consideration of the fabric properties in surface development. The proposed method was demonstrated to effectively realize the pattern generation of 3D sample clothes with fabric properties without damaging the garment samples, and to be suitable for different clothing styles and fabrics. Compared with traditional methods, the proposed method has higher accuracy (2.21% higher on average) and better stability.
{"title":"An undamaged pattern generation method from 3D scanned garment\u0000sample based on finite element approach","authors":"JIA-ZHEN Chen, ZI-YI Guo, Tao Li, Lei Du, F. Zou","doi":"10.35530/it.074.01.202185","DOIUrl":"https://doi.org/10.35530/it.074.01.202185","url":null,"abstract":"The purpose of this study is to propose a new method to achieve pattern generation from garment sample without\u0000damage. The non-contact three-dimensional (3D) scanner was employed to get the point cloud data of garment\u0000samples. The Bowyer-Watson algorithm was used to implement Delaunay triangulation for surface reconstruction. The\u0000finite element (FE) approach was employed to achieve the consideration of the fabric properties in surface development.\u0000The proposed method was demonstrated to effectively realize the pattern generation of 3D sample clothes with fabric\u0000properties without damaging the garment samples, and to be suitable for different clothing styles and fabrics. Compared\u0000with traditional methods, the proposed method has higher accuracy (2.21% higher on average) and better stability.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47731290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-28DOI: 10.35530/it.074.01.202199
Tao Li, Yexin Lyu, Ling Ma, Younglun Xie, Fengyuan Zou
The automatic recognition of garment flat information has been widely researched through computer vision. However, the unapparent visual feature and low recognition accuracy pose serious challenges to the application. Herein, inspired by multi-object instance segmentation, the method of mask region convolutional neural network (Mask R-CNN) for garment flat multi-component is proposed in this paper. The steps include feature enhancement, attribute annotation, feature extraction, and bounding box regression and recognition. First, the Laplacian was employed to enhance the image feature, and the Polygon annotated component attributes to reduce the interaction interference. Next, the ResNet was applied to realize identity mapping to characterize redundant information of components. Finally, the feature map was entered into two branches to achieve bounding box regression and recognition. The results demonstrated that the proposed method could realize multi-component recognition effectively. Compared with the unenhanced feature, the mAP increased by 2.27%, reaching 97.87%, and the average F1 was 0.958. Compared to VGGNet and MobileNet, the ResNet backbone used for Mask R-CNN could improve the mAP by 11.55%. Mask R-CNN was more robust than the state-of-the-art methods and more suitable for garment flat multi-component recognition.
{"title":"Research on garment flat multi-component recognition based on\u0000Mask R -CNN","authors":"Tao Li, Yexin Lyu, Ling Ma, Younglun Xie, Fengyuan Zou","doi":"10.35530/it.074.01.202199","DOIUrl":"https://doi.org/10.35530/it.074.01.202199","url":null,"abstract":"The automatic recognition of garment flat information has been widely researched through computer vision. However,\u0000the unapparent visual feature and low recognition accuracy pose serious challenges to the application. Herein, inspired\u0000by multi-object instance segmentation, the method of mask region convolutional neural network (Mask R-CNN) for\u0000garment flat multi-component is proposed in this paper. The steps include feature enhancement, attribute annotation,\u0000feature extraction, and bounding box regression and recognition. First, the Laplacian was employed to enhance the\u0000image feature, and the Polygon annotated component attributes to reduce the interaction interference. Next, the ResNet\u0000was applied to realize identity mapping to characterize redundant information of components. Finally, the feature map\u0000was entered into two branches to achieve bounding box regression and recognition. The results demonstrated that the\u0000proposed method could realize multi-component recognition effectively. Compared with the unenhanced feature, the\u0000mAP increased by 2.27%, reaching 97.87%, and the average F1 was 0.958. Compared to VGGNet and MobileNet, the\u0000ResNet backbone used for Mask R-CNN could improve the mAP by 11.55%. Mask R-CNN was more robust than the\u0000state-of-the-art methods and more suitable for garment flat multi-component recognition.","PeriodicalId":13638,"journal":{"name":"Industria Textila","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47997312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：