At present,convolutional neural networks(CNNs)and transformers surpass humans in many situations(such as face recognition and object classification),but do not work well in identifying fibers in textile surface images...At present,convolutional neural networks(CNNs)and transformers surpass humans in many situations(such as face recognition and object classification),but do not work well in identifying fibers in textile surface images.Hence,this paper proposes an architecture named FiberCT which takes advantages of the feature extraction capability of CNNs and the long-range modeling capability of transformer decoders to adaptively extract multiple types of fiber features.Firstly,the convolution module extracts fiber features from the input textile surface images.Secondly,these features are sent into the transformer decoder module where label embeddings are compared with the features of each type of fibers through multi-head cross-attention and the desired features are pooled adaptively.Finally,an asymmetric loss further purifies the extracted fiber representations.Experiments show that FiberCT can more effectively extract the representations of various types of fibers and improve fiber identification accuracy than state-of-the-art multi-label classification approaches.展开更多
Two-dimensional(2D)MXenes have emerged as an archetypical layered material combining the properties of an organic-inorganic hybrid offering materials sustainability for a range of applications.Their surface functional...Two-dimensional(2D)MXenes have emerged as an archetypical layered material combining the properties of an organic-inorganic hybrid offering materials sustainability for a range of applications.Their surface functional groups and the associated chemical properties'tailorability through functionalizing MXenes with other materials as well as hydrophilicity and high conductivity enable them to be the best successor for various applications in textile industries,especially in the advancement of smart textiles and remediation of textile wastewater.MXene-based textile composite performs superb smartness in high-performance wearables as well as in the reduction of textile dyes from wastewater.This article critically reviews the significance of MXenes in two sectors of the textile industry.Firstly,we review the improvement of textile raw materials such as fiber,yarn,and fabric by using MXene as electrodes in supercapacitors,pressure sensors.Secondly,we review advancements in the removal of dyes from textile wastewater utilizing MXene as an absorbent by the adsorption process.MXene-based textiles demonstrated superior strength through the strong bonding between MXene and textile structures as well as the treatment of adsorbate by adsorbent(MXene in the adsorption process).We identify critical gaps for further research to enable their real-life applications.展开更多
基金National Natural Science Foundation of China(No.61972081)Fundamental Research Funds for the Central Universities,China(No.2232023Y-01)Natural Science Foundation of Shanghai,China(No.22ZR1400200)。
文摘At present,convolutional neural networks(CNNs)and transformers surpass humans in many situations(such as face recognition and object classification),but do not work well in identifying fibers in textile surface images.Hence,this paper proposes an architecture named FiberCT which takes advantages of the feature extraction capability of CNNs and the long-range modeling capability of transformer decoders to adaptively extract multiple types of fiber features.Firstly,the convolution module extracts fiber features from the input textile surface images.Secondly,these features are sent into the transformer decoder module where label embeddings are compared with the features of each type of fibers through multi-head cross-attention and the desired features are pooled adaptively.Finally,an asymmetric loss further purifies the extracted fiber representations.Experiments show that FiberCT can more effectively extract the representations of various types of fibers and improve fiber identification accuracy than state-of-the-art multi-label classification approaches.
基金the University Malaysia Pahang for the financial aid providing the grants(Nos.RDU 213308 and RDU 192207).
文摘Two-dimensional(2D)MXenes have emerged as an archetypical layered material combining the properties of an organic-inorganic hybrid offering materials sustainability for a range of applications.Their surface functional groups and the associated chemical properties'tailorability through functionalizing MXenes with other materials as well as hydrophilicity and high conductivity enable them to be the best successor for various applications in textile industries,especially in the advancement of smart textiles and remediation of textile wastewater.MXene-based textile composite performs superb smartness in high-performance wearables as well as in the reduction of textile dyes from wastewater.This article critically reviews the significance of MXenes in two sectors of the textile industry.Firstly,we review the improvement of textile raw materials such as fiber,yarn,and fabric by using MXene as electrodes in supercapacitors,pressure sensors.Secondly,we review advancements in the removal of dyes from textile wastewater utilizing MXene as an absorbent by the adsorption process.MXene-based textiles demonstrated superior strength through the strong bonding between MXene and textile structures as well as the treatment of adsorbate by adsorbent(MXene in the adsorption process).We identify critical gaps for further research to enable their real-life applications.