检索结果-内蒙古大学图书馆

D2SFormer: Dual attention-dynamic bidirectional Transformer for semantic segmentation of urban remote sensing images

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 12248-12262页

作者： Yan, Yi Li, Jiafeng Zhang, Jing Wang, Liuqian Zhuo, Li Beijing University of Technology School of Information Science and Technology Chaoyang Beijing100124 China Beijing University of Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System China

Semantic segmentation of urban remote sensing images (URSIs) is crucial for the development of smart cities and the acceleration of urban information. In view of the characteristics of URSIs with intricate backgrounds, dense distributions of feature objects, and varying object scales, we propose a dual attention-dynamic bidirectional Transformer (D2SFormer) for semantic segmentation of URSIs. First, a dual attention Transformer encoder has a global branch of cross-shaped window self-attention (CWSA) and a local branch of convolutional channel attention block (CCAB), which helps to enhance pixel-level semantic discrimination. Then, CCAB combines the local sensitivity of Conv with the non-local modeling capability of ECA for overlapping region resolution. Finally, the decoder dynamically fuses multiscale features to reconstruct boundaries and suppress artifacts by using a bidirectional feature pyramid network (BiFPN) and dysampling. Experimental results demonstrate that our D2SFormer achieves mIoUs of 87.84%, 81.24%, and 53.19% on the Potsdam, Vaihingen, and LoveDA datasets, respectively, which is highly competitive with existing methods. © 2008-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Recycling the spent electronic materials to construct a highperformance Cu_(1.94)S/ZnS heterostructure anode of sodium-ion batteries

引用

Nano Research 2024年第5期17卷 4006-4015页

作者： Xiaowei He Tianshuai Wang Lidong Tian Qiuyu Zhang School of Chemistry and Chemical Engineering Xi’an Key Laboratory of Functional Organic Porous MaterialsKey Laboratory of Special Functional and Smart Polymer Materials of Ministry of Industry and Information TechnologyNorthwestern Polytechnical UniversityXi’an 710072China Chongqing Science and Technology Innovation Center of Northwestern Polytechnical University Chongqing 401135China College of Materials Science and Engineering Xi’an University Science and TechnologyXi’an 710054China

Heterostructure engineering by coupling different nanocrystals has received extensive attention because it can enhance the reaction kinetics of the anode of sodium-ion batteries(SIBs).However,constructing high-quality heterostructure anode materials through green and environmentally friendly methods remains a ***,we have proposed a simple one-step method by recycling the electronic waste metal materials to synthesize the Cu_(1.94)S/ZnS heterostructure *** with the experimental analysis and first principle calculations,we find that the synergistic effect of different components in heterostructure structures can significantly enhance the reversible capacity and rate performance of anode *** on the constructed Cu_(1.94)S/ZnS anode,we obtain a superior reversible capacity of 440 mAh·g^(-1) at 100 mA·g^(-1) and 335 mAh·g^(-1) after 3000 cycles at 2000 mA·g^(-1).Our work sheds new light on designing high-rate and capacity anodes for SIBs through the greenness synthesis method.

关键词： sodium-ion batteries heterostructure reversible capacity environmentally synthesis process

来源：评论

学校读者我要写书评

暂无评论

Co-regulation effect of solvation and interface of pyridine derivative enabling highly reversible zinc anode

引用

Journal of Materials science & technology 2025年第1期204卷 1-9页

作者： Binrui Xu Guangbin Wang Yong Liu Quanan Li Fengzhang Ren Jianmin Ma School of Information Engineering Henan University of Science and TechnologyLuoyang 471023China School of Materials Science and Engineering Provincial and Ministerial Coconstruction of Collaborative Innovation Center for Non-ferrous Metal New Materials and Advanced Processing TechnologyHenan University of Science and TechnologyLuoyang 471023China Longmen laboratory Luoyang 471000China School of Chemistry Tiangong UniversityTianjin 300387China

The poor reversibility and stability of Zn anodes greatly restrict the practical application of aqueous Zn-ion batteries(AZIBs),resulting from the uncontrollable dendrite growth and H_(2)O-induced side reactions during *** additive modification is considered one of the most effective and simplest methods for solving the aforementioned ***,the pyridine derivatives(PD)including 2,4-dihydroxypyridine(2,4-DHP),2,3-dihydroxypyridine(2,3-DHP),and 2-hydroxypyrdine(2-DHP),were em-ployed as novel electrolyte additives in ZnSO_(4)*** density functional theory calculation and experimental findings demonstrated that the incorporation of PD additives into the electrolyte effectively modulates the solvation structure of hydrated Zn ions,thereby suppressing side reactions in ***-ditionally,the adsorption of PD molecules on the zinc anode surface contributed to uniform Zn deposi-tion and dendrite growth ***,a 2,4-DHP-modified Zn/Zn symmetrical cell achieved an extremely long cyclic stability up to 5650 h at 1 mA cm^(-2).Furthermore,the Zn/NH_(4)V_(4)O_(10)full cell with 2,4-DHP-containing electrolyte exhibited an outstanding initial capacity of 204 mAh g^(-1),with a no-table capacity retention of 79%after 1000 cycles at 5 A g^(-1).Hence,this study expands the selection of electrolyte additives for AZIBs,and the working mechanism of PD additives provides new insights for electrolyte modification enabling highly reversible zinc anode.

关键词： Zn anode Pyridine derivative Electrolyte additive Solvation regulation Interface modification

来源：评论

学校读者我要写书评

暂无评论

Femtosecond laser-induced periodic surface structures on hard and brittle materials

引用

science China(Technological sciences) 2024年第1期67卷 19-36页

作者： ZHAO GuoXu WANG Gong LI YunFei WANG Lei LIAN YuDong YU Yu ZHAO Hui WANG YuLei LU ZhiWei Center for Advanced Laser Technology Hebei University of TechnologyTianjin 300401China Hebei Key Laboratory of Advanced Laser Technology and Equipment Tianjin 300401China State Key Laboratory of Integrated Optoelectronics College of Electronic Science and EngineeringJilin UniversityChangchun 130012China School of Information and Electrical Engineering Hebei University of EngineeringHandan 056038China

Hard and brittle materials have high hardness,excellent optical stability,chemical stability,and high thermal ***,they have huge application potential in various fields,such as optical components,substrate materials,and quantum information,especially under harsh conditions,such as high temperatures and high *** laser direct writing technology has greatly promoted the development of femtosecond laser-induced periodic surface structure(Fs-LIPSS or LIPSS by a femtosecond laser)applications of hard and brittle materials due to its high precision,controllability,and three-dimensional processing *** far,LIPSSs have been widely used in material surface treatment,optoelectronic devices,and ***,a consensus has not been reached regarding the formation mechanism of LIPSSs on hard and brittle *** this paper,three widely accepted LIPSS formation mechanisms are introduced,and the characteristics and applications of LIPSSs on diamonds,silicon,silicon carbide,and fused silica surfaces in recent years are *** addition,the application prospects and challenges of LIPSSs on hard and brittle materials by a femtosecond laser are discussed.

关键词： laser-induced periodic surface structures hard and brittle materials femtosecond laser

来源：评论

学校读者我要写书评

暂无评论

ViGT: proposal-free video grounding with a learnable token in the transformer

引用

science China(information sciences) 2023年第10期66卷 196-212页

作者： Kun LI Dan GUO Meng WANG School of Computer Science and Information Engineering Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Intelligent Interconnected Systems Laboratory of Anhui Province Institute of Artificial Intelligence Hefei Comprehensive National Science Center

The video grounding(VG) task aims to locate the queried action or event in an untrimmed video based on rich linguistic descriptions. Existing proposal-free methods are trapped in the complex interaction between video and query, overemphasizing cross-modal feature fusion and feature correlation for VG. In this paper, we propose a novel boundary regression paradigm that performs regression token learning in a transformer. Particularly, we present a simple but effective proposal-free framework, namely video grounding transformer(ViGT), which predicts the temporal boundary using a learnable regression token rather than multi-modal or cross-modal features. In ViGT, the benefits of a learnable token are manifested as follows.(1) The token is unrelated to the video or the query and avoids data bias toward the original video and query.(2) The token simultaneously performs global context aggregation from video and query ***, we employed a sharing feature encoder to project both video and query into a joint feature space before performing cross-modal co-attention(i.e., video-to-query attention and query-to-video attention) to highlight discriminative features in each modality. Furthermore, we concatenated a learnable regression token [REG] with the video and query features as the input of a vision-language transformer. Finally, we utilized the token [REG] to predict the target moment and visual features to constrain the foreground and background probabilities at each timestamp. The proposed ViGT performed well on three public datasets:ANet-Captions, TACoS, and YouCookⅡ. Extensive ablation studies and qualitative analysis further validated the interpretability of ViGT.

关键词： video grounding temporal sentence grounding boundary regression token learning proposal-free

来源：评论

学校读者我要写书评

暂无评论

Modulation of inactive Li_(2)O via iodinated MOF nanocapsules interfacial transformation engineering for high-performance solid electrolyte interphase

引用

Journal of Energy Chemistry 2025年第5期104卷 482-493页

作者： Xingxing Zhang Qingmei Su Gaohui Du Bingshe Xu Xuehan Hou Xiaowei Yang Weihao Shi Zhuo Chen Yang Shi Yujie Lv Wenhuan Huang School of Materials Science&Engineering Shaanxi University of Science and TechnologyXi’an 710021ShaanxiChina Key Laboratory of Chemical Additives for China National Light Industry College of Chemistry and Chemical EngineeringShaanxi University of Science and TechnologyXi’an 710021ShaanxiChina Materials Institute of Atomic and Molecular Science School of Physics&Information ScienceShaanxi University of Science and TechnologyXi’an 710021ShaanxiChina School of Chemistry and Chemical Engineering Shanghai Jiaotong UniversityShanghai 200240China

Lithium(Li)deposition and nucleation at solid electrolyte interphase(SEI)is the main origin for the capacity decay in Li metal batteries(LMBs).SEI conversion with enhanced electrochemical and mechanical properties is an effective approach to achieve uniform nucleation of Li^(+)and stabilize the lithium metal ***,complex interfacial reaction mechanisms and interface compatibility issues hinder the development of SEI conversion strategies for stabilizing lithium metal ***,we presented the release of I_(3)^(-)in–NH_(2)-modified metal–organic frameworks for a Li metal surface SEI phase conversion ***–NH_(2)group in MOF pores induced the formation of I_(3)^(-)from I_(2),which was further spontaneously reacted with inactive Li_(2)O transforming into high-performance LiI and LiIO_(3)***,theoretical calculation provided deeply insight into the unique reconstructed interfacial formation and electrochemical mechanism of rich LiI and LiIO_(3)*** a result,the Li^(+)deposition and nucleation were improved,facilitating the transport kinetics of Li^(+)and inhibiting the growth of lithium *** assembled solid-state Li||LiFePO_(4)full cells exhibited superior long-term stability of 800 cycles and high Coulombic efficiency(>99%),Li||LiNi_(0.8)Co_(0.1)Mn_(0.1)O_(2)pouch cell also displayed superior practical performance over 200 cycles at 2 C,high loading of 5 mg cm^(-2)and safety *** innovative SEI design strategy promotes the development of high-performance solid-state Li metal batteries.

关键词： Solid electrolyte interphase(SEI) SEI phase conversion MOF Nano-capsule Solid-state electrolytes Li metal battery

来源：评论

学校读者我要写书评

暂无评论

Towards optimized tensor code generation for deep learning on sunway many-core processor

引用

Frontiers of Computer science 2024年第2期18卷 1-15页

作者： Mingzhen LI Changxi LIU Jianjin LIAO Xuegui ZHENG Hailong YANG Rujun SUN Jun XU Lin GAN Guangwen YANG Zhongzhi LUAN Depei QIAN State Key Laboratory of Software Development Environment Beijing 100191China School of Computer Science and Engineering Beihang UniversityBeijing 100191China National University of Singapore Singapore 119077Singapore State Key Laboratory of Mathematical Engineering and Advanced Computing Wuxi 214000China Science and Technology on Special System Simulation Laboratory Beijing Simulation Center Beijing 100854China Department of Computer Science and Technology Tsinghua UniversityBeijing 100084China

The flourish of deep learning frameworks and hardware platforms has been demanding an efficient compiler that can shield the diversity in both software and hardware in order to provide application *** the existing deep learning compilers,TVM is well known for its efficiency in code generation and optimization across diverse hardware *** the meanwhile,the Sunway many-core processor renders itself as a competitive candidate for its attractive computational power in both scientific computing and deep learning *** paper combines the trends in these two ***,we propose swTVM that extends the original TVM to support ahead-of-time compilation for architecture requiring cross-compilation such as *** addition,we leverage the architecture features during the compilation such as core group for massive parallelism,DMA for high bandwidth memory transfer and local device memory for data locality,in order to generate efficient codes for deep learning workloads on *** experiment results show that the codes generated by swTVM achieve 1.79x improvement of inference latency on average compared to the state-of-the-art deep learning framework on Sunway,across eight representative *** work is the first attempt from the compiler perspective to bridge the gap of deep learning and Sunway processor particularly with productivity and efficiency in *** believe this work will encourage more people to embrace the power of deep learning and Sunwaymany-coreprocessor.

关键词： sunway processor deep learning compiler code generation performance optimization

来源：评论

学校读者我要写书评

暂无评论

Geometry-Based Stochastic MIMO Channel Model for Near-Field and Far-Field Scenarios of Integrated Sensing and Communications

引用

IEEE Transactions on Vehicular technology 2025年第5期74卷 6928-6940页

作者： Jin, Yunwei He, Ruisi Ai, Bo Yuan, Yuan Niu, Yong Zhang, Haoxiang Beijing Jiaotong University State Key Laboratory of Advanced Rail Autonomous Operation School of Electronics and Information Engineering Frontiers Science Center for Smart High-speed Railway System Beijing100044 China Ministry of Industry and Information Technology China Academy of Industrial Internet Beijing100804 China

Integrated sensing and communication (ISAC) has become a promising technology in current sixth-generation (6G) wireless communications due to its ability to support both communication and sensing. Existing works regarding ISAC channels mainly focus on far-field propagation conditions based on plane wave assumption. However, with the deployment of ultra-massive multiple-input multiple-output (MIMO), the far-field assumption may no longer hold, as the communication and sensing propagation distances distances may be comparable to the array aperture size. To address this issue, we propose a three-dimensional MIMO channel model that simultaneously considers near-field and far-field propagation conditions in ISAC environments. Additionally, we distinguish the communication and sensing propagation environments by employing different effective scatterer distributions. In the model, the sensing channel is divided into target sensing and environmental sensing components, whereas the communication channel is divided into line-of-sight and non-line-of-sight components. The weighted sum of each component in communication and sensing channels results in accurate channel representations. Based on the proposed model, we derive and thoroughly investigate space-time-frequency correlation function, normalized absolute error function, and channel capacity. A key observation is that the channel correlation strongly depends on scatterer distribution. In certain conditions, the sensing channel can effectively capture the behavior of the communication channel. Error function results indicate that the proposed channel model achieves fairly high accuracy in ultra-massive MIMO scenarios. These findings provide essential support for the development of ISAC systems in future 6G wireless communications. © 1967-2012 IEEE.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

BayProNet: Advancing Few-Shot Learning with Bayesian Prototype Network 24

BayProNet: Advancing Few-Shot Learning with Bayesian Prototy...

引用

8th International Conference on Computer science and Artificial Intelligence, CSAI 2024

作者： Gao, Cheng Lang Miao, Jiang Hui, Chen Hong Science and Technology on Information Systems Engineering Laboratory National University of Defense Technology HuNan ChangSha China

ISBN: (纸本)9798400718182

Text classification is a challenging task in the field of Natural Language Processing (NLP), and significant progress has been made using deep learning methods. Traditional deep-learning approaches for text classification require large amounts of data, thereby performing suboptimally in few-shot data scenarios. To alleviate this bottleneck, this study proposes a Bayesian Prototypical Network with Hybrid Attention (BayProNet) to improve the accuracy of few-shot text classification. Experimental results on the Clinc150 dataset show that the model significantly outperforms existing SOTA baseline models in terms of classification accuracy and also exhibits strong stability when handling noisy data. This research validates the effectiveness and superiority of Bayesian Prototypical Networks in few-shot text classification tasks. © 2024 Copyright held by the owner/author(s).

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Closed-loop scheduling optimization strategy based on particle swarm optimization with niche technology and soft sensor method of attributes-applied to gasoline blending process

引用

Chinese Journal of Chemical engineering 2023年第9期61卷 43-57页

作者： Jian Long Kai Deng Renchu He Key Laboratory of Smart Manufacturing in Energy Chemical Process Ministry of EducationEast China University of Science and TechnologyShanghaiChina Engineering Research Center of Process System Engineering Ministry of EducationEast China University of Science and TechnologyShanghai 200237China

Gasoline blending scheduling optimization can bring significant economic and efficient benefits to ***,the optimization model is complex and difficult to build,which is a typical mixed integer nonlinear programming(MINLP)*** the large scale of the MINLP model,in order to improve the efficiency of the solution,the mixed integer linear programming-nonlinear programming(MILP-NLP)strategy is used to solve the *** paper uses the linear blending rules plus the blending effect correction to build the gasoline blending model,and a relaxed MILP model is constructed on this *** particle swarm optimization algorithm with niche technology(NPSO)is proposed to optimize the solution,and the high-precision soft-sensor method is used to calculate the deviation of gasoline attributes,the blending effect is dynamically corrected to ensure the accuracy of the blending effect and optimization results,thus forming a prediction-verification-reprediction closed-loop scheduling optimization strategy suitable for engineering *** optimization result of the MILP model provides a good initial *** fixing the integer variables to the MILPoptimal value,the approximate MINLP optimal solution can be obtained through a NLP *** above solution strategy has been successfully applied to the actual gasoline production case of a refinery(3.5 million tons per year),and the results show that the strategy is effective and *** optimization results based on the closed-loop scheduling optimization strategy have higher *** with the standard particle swarm optimization algorithm,NPSO algorithm improves the optimization ability and efficiency to a certain extent,effectively reduces the blending cost while ensuring the convergence speed.

关键词： Blend Optimization algorithm Neural networks Particle swarm optimization Mixed integer programming

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：