作者机构:
[Xie, Fuqiang; Guo, Lei; Yang, Bin] Univ South China, Sch Elect Engn, Hengyang, Peoples R China.
会议名称:
8th International Conference on Hydraulic and Civil Engineering - Deep Space Intelligent Development and Utilization Forum (ICHCE)
会议时间:
NOV 25-27, 2022
会议地点:
Xian, PEOPLES R CHINA
会议主办单位:
[Guo, Lei;Xie, Fuqiang;Yang, Bin] Univ South China, Sch Elect Engn, Hengyang, Peoples R China.
关键词:
component;Deep learning;Lane detection;non-local attention;frame-similarity loss
摘要:
Lane detection plays an important role in autonomous driving. For video instance lane detection, both global spatial and temporal information is significantly important. However, the global spatial features and the temporal features are not been well exploited in recent studies. In this work, we address the video instance lane detection task by capturing global context based on non-local attention network. Specifically, we designed a twin non-local attention network to extract long-range dependencies along the spatial and temporal dimensions, respectively. Meanwhile, the global spatial and temporal features can be adaptively fused by gating mechanisms for better results. In addition, the frame-similarity loss is proposed to further exploit the information of adjacent frames. The experimental results on the video instance lane detection (VIL-100) dataset verify that our method achieves better results compared with other comparison methods. Ablation experiments further demonstrate the effectiveness of each sub-module.
通讯机构:
[Aixiang Wei] G;Guangdong Provincial Key Laboratory of Functional Soft Condensed Matter, School of Integrated Circuits, Guangdong University of Technology, Guangzhou 510006, Guangdong, China<&wdkj&>School of Information Science,Guangzhou Xinhua University, Dongguan 523133, Guangdong, China
摘要:
Rhenium diselenide (ReSe2) has attracted great interest due to its unique anisotropic structure and unusual in-plane anisotropic electrical and optical properties. However, efficient fabrication of large-area and high-quality 2D ReSe2 continuous films has become an increasingly important challenge. In this work, centimeter-scale 2D ReSe2 continuous films with the layer number varying from monolayer to 12 layers were successfully grown on a mica substrate using our space-confined CVD system via changing the position of the substrate. The fluorescence quenching effect and surface enhanced Raman scattering (SERS) effect on 2D ReSe2 films with different layer numbers were investigated using rhodamine 6G (R6G) dye molecules as a Raman probe. The large-area ReSe2 films show the layer-number-dependent nature of the SERS effect and a robust suppression effect of fluorescence. Our work explores the practical application of 2D ReSe2 films for molecular detection via the SERS technique.
作者机构:
[Jiang, Shangting; Li, Ye; Wang, Xinlin] School of Nuclear Science and Technology, University of South China, No. 28, Changsheng West Road, Hunan, Hengyang;421001, China;[Chen, Zhiyong; Zhu, Weihua; Wu, Qinmao; He, Hongyu] Hunan Province Key Laboratory for Ultra-Fast Micro/Nano Technology and Advanced Laser Manufacture, School of Electrical Engineering, University of South China, No. 28, Changsheng West Road, Hunan, Hengyang;[Jiang, Shangting; Li, Ye; Chen, Zhiyong; Zhu, Weihua; Wu, Qinmao; He, Hongyu; Wang, Xinlin] 421001, China;[Wang, Xinlin] 421001, China<&wdkj&>Hunan Province Key Laboratory for Ultra-Fast Micro/Nano Technology and Advanced Laser Manufacture, School of Electrical Engineering, University of South China, No. 28, Changsheng West Road, Hunan, Hengyang
通讯机构:
[Ye Li; Xinlin Wang] S;School of Nuclear Science and Technology, University of South China , No. 28, Changsheng West Road, Hengyang City, Hunan 421001, China
作者机构:
[Luo, Xiao-Qing; Zhu, Weihua; Xu, Xiaofeng; Liu, Qinke; Chen, Zhiyong; Wang, Xin-Lin] Univ South China, Sch Elect Engn, Hunan Prov Key Lab Ultrafast Micro Nano Technol &, Hengyang 421001, Peoples R China.;[Li, Yan] Univ South China, Sch Nucl Sci & Technol, Hengyang 421001, Peoples R China.;[Liu, Wuming] Chinese Acad Sci, Inst Phys, Be?ing Natl Lab Condensed Matter Phys, Be?ing 100190, Peoples R China.;[Wang, Xin-Lin] Univ South China, Sch Mech Engn, Hengyang 421001, Peoples R China.
通讯机构:
[Xiao-Qing Luo; Xin-Lin Wang] H;Hunan Province Key Laboratory for Ultra-Fast Micro/Nano Technology and Advanced Laser Manufacture, School of Electrical Engineering, University of South China, Hengyang 421001, China<&wdkj&>School of Mechanical Engineering, University of South China, Hengyang 421001, China<&wdkj&>Hunan Province Key Laboratory for Ultra-Fast Micro/Nano Technology and Advanced Laser Manufacture, School of Electrical Engineering, University of South China, Hengyang 421001, China<&wdkj&>Authors to whom correspondence should be addressed.
摘要:
Abstract: Fano resonances that feature strong field enhancement in the narrowband range have motivated extensive studies of light–matter interactions in plasmonic nanomaterials. Optical metasurfaces that are subject to different mirror symmetries have been dedicated to achieving nanoscale light manipulation via plasmonic Fano resonances, thus enabling advantages for high-sensitivity optical sensing and optical switches. Here, we investigate the plasmonic sensing and switches enriched by tailorable multiple Fano resonances that undergo in-plane mirror symmetry or asymmetry in a hybrid rotational misalignment metasurface, which consists of periodic metallic arrays with concentric C-shaped- and circular-ring-aperture unit cells. We found that the plasmonic double Fano resonances can be realized by undergoing mirror symmetry along the X-axis. The plasmonic multiple Fano resonances can be tailored by adjusting the level of the mirror asymmetry along the Z-axis. Moreover, the Fano-resonance-based plasmonic sensing that suffer from mirror symmetry or asymmetry can be implemented by changing the related structural parameters of the unit cells. The passive dual-wavelength plasmonic switches of specific polarization can be achieved within mirror symmetry and asymmetry. These results could entail benefits for metasurface-based devices, which are also used in sensing, beam-splitter, and optical communication systems. Keywords: Fano resonance; metasurface; plasmonic sensing; plasmonic switch
作者机构:
[Zhang, XZ; Zhang, Xiaozhi; Zhou, Liu] Univ South China, Sch Elect Engn, Hengyang 421001, Peoples R China.;[Zhu, Minjie; Ouyang, Lijun; Ouyang, Yan; Xiong, Dongping; Ouyang, LJ] Univ South China, Sch Comp Software, Hengyang 421001, Peoples R China.
会议名称:
14th International Conference on Graphics and Image Processing (ICGIP)
会议时间:
OCT 21-23, 2022
会议地点:
ELECTR NETWORK
会议主办单位:
[Zhou, Liu;Zhang, Xiaozhi] Univ South China, Sch Elect Engn, Hengyang 421001, Peoples R China.^[Zhu, Minjie;Xiong, Dongping;Ouyang, Lijun;Ouyang, Yan] Univ South China, Sch Comp Software, Hengyang 421001, Peoples R China.
会议论文集名称:
Proceedings of SPIE
关键词:
Magnetic resonance imaging (MRI);fast MRI;deep learning;Non-Local;remote dependencies
摘要:
As an advanced medical imaging technology, magnetic resonance imaging (MRI) has great advantages and application potentials in medical clinical diagnosis. However, since the long scanning time and the artifacts caused by patient movements, the imaging results are always not satisfactory. Therefore, accelerating MRI and improving the imaging quality are the key problems. In this work, we propose a novel deep network that combines the U-net architecture with non-local attention blocks for MRI reconstruction. We employ the U-net to construct the basic network. The non-local attention is exploited to capture the remote dependencies in MRI images which calculates the weighted average of the remaining multiple location features as the value of the response location. The U-net has limitations in capturing long-term dependencies, however, the non-local attention can solve this problem well. Furthermore, we develop the residual module to better retain the detail information. The proposed model is compared with some recent leading MRI reconstruction methods, including the state-of-the-art deep learning-based methods. Compared with these methods, the proposed residual non-local attention network provides superior MRI reconstruction results and retains better perceptual image details.
通讯机构:
[He, HY ] U;Univ South China, Sch Elect Engn, Hengyang 421001, Peoples R China.;Peking Univ, Sch Elect & Comp Engn, Shenzhen Grad Sch, Shenzhen 518005, Peoples R China.;Guangdong Univ Technol, Sch Automation, Guangzhou 510006, Peoples R China.
关键词:
Thin-film transistor (TFT);Drain current model;Temperature characteristics;Trap states
摘要:
A new analytical surface-potential-based drain current model is presented for the organic thin-film transistors (TFTs) when carriers are confined in two dimensions. Following the carrier multiple trapping and release (MTR) conduction theory, i.e., the assumption that the trapped carrier concentration is much higher than the free carrier concentration, the model is developed. The presented model can account for the linear regime and saturation regime by a single formulation. The calculated results of the presented model are verified by the available experimental drain current considering the temperature characteristics. Comparing with the previous model using the variable range hopping and percolation (VRH) conduction theory, although the presented model and the previous model are similar in mathematics, the presented model is more efficient to estimate the density of trap states for the organic TFTs.
摘要:
A hybrid system with jointed battery and PEMFC is popular and of great potential in New Energy Vehicle (NEV) application. However, reliability and efficiency remain to be improved for commercial products. To reflect the complicated physics inside the proton exchange membrane fuel cell (PEMFC), the PEMFC model consisting of inner muti-physics process and other accessories was built, then a complete hybrid system was established when a matched battery, DC/DC, regenerative braking were taken into consideration. Based on the above model, the stack state and system performance under standard cycle for heavy duty vehicleCWTVC were obtained. According to the simulation results, fuel cell states such as pressure, water content and voltage suffers severe oscillation with external load, especially in the highway cycle. Membrane electrode assembly (MEA) suffers from pressure impact with average value of more than 24 kPa in highway cycle. In the aspect of relative humidity, the PEMFC stack is most threatened in road cycle. As for the hybrid system, its efficiency and state of charge (SOC) fluctuation perform worst in urban cycle and road cycle respectively, while its highest efficiency occurs in road test. Operating mode of fuel cell has influence on hybrid system. When 3-level mode of fuel cell output was applied, the efficiency increased to its peak value at medium level of 28 kW and then declined gradually. H2 consumption had an opposite trend compared to efficiency. In the aspect of battery SOC, it declines in operating process and its fluctuations decreases when medium level got bigger. The 3-level mode and 4-level mode were compared using this model. It can be concluded that although 3-level mode performs slightly better in hybrid system efficiency, H2 consumption, pressure impact, it does not have absolute advantage over 4-level mode in other indicators. (c) 2021 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
关键词:
6G;6G mobile communication;Confident Information Coverage Model(CIC);Energy consumption;Green IoT;Green products;Internet of Things;Monitoring;Network Lifetime;Reinforcement Learning;Sensors;Sleep Scheduling;Uranium
摘要:
The Internet of Things (IoT) enabled by 6G increases the number of devices and users exponentially. 6G IoT will cover the overall domain of the world. Due to the limited resources in the IoT environment, greener and smarter networks are indispensable for the sustainable development of the 6G IoT. In this paper, the issue of improving resource efficiency and extending network lifetime is studied. We propose a novel Confidence Information Coverage (CIC) node sleep scheduling algorithm based on reinforcement learning (CICRL). In CICRL, collaborative intelligence is achieved through Q-learning to meet the coverage rate with the least active nodes, thus balancing the energy consumption and prolonging the network lifetime. Compared with the Coordination Algorithm based on Reinforcement Learning (COORD), Low-Energy Adaptive Clustering Hierarchy (LEACH) and Sleep Scheduling Approach based on Learning Automata (PCLA), the simulation results demonstrate that the proposed algorithm satisfies coverage with fewer active nodes and improves the network lifetime substantially.