检索结果-内蒙古大学图书馆

Towards Gradient Equalization and Feature Diversification for Long-Tailed Multi-Label Image Recognition

IEEE Transactions on Multimedia 2025年 27卷 3489-3500页

作者： Chen, Zhao-Min Cui, Quan Zhang, Xiaoqin Deng, Ruoxi Xia, Chaoqun Lu, Shijian Nanjing University National Key Laboratory for Novel Software Technology Nanjing210093 China Waseda University Graduate School of Information Production and Systems Fukuoka169-8050 Japan Wenzhou University College of Computer Science and Artificial Intelligence Wenzhou325060 China Nanyang Technological University School of Computer Science and Engineering 639798 Singapore

Multi-label image recognition with convolutional neural networks has achieved remarkable progress in the past few years. However, most existing multi-label image recognition methods suffer from the long-tailed data distribution problem, i.e., head categories occupy most training samples, while tailed classes have few samples. This work firstly studies the influence of long-tailed data distribution on existing multi-label image recognition methods. Based on this, two crucial issues of the existing methods are identified: 1) severe gradient imbalance between head and tailed categories, even though re-balancing strategies are adopted;2) the lack of diversity of tail category training samples. To tackle the first issue, this paper proposes a group sampling strategy to create group- wise balanced data distribution. Meanwhile, a dynamic gradient balancing loss is proposed to equalize the gradient for all categories. To tackle the second issue, this paper proposes a diversity enhancement module to fuse the information across all categories, preventing the network from overfitting tail classes. Furthermore, it also balances the gradient, promoting the discriminability of learned classifiers. Our method significantly outperforms the baseline method and achieves competitive performance with state-of-the-art methods on VOC-LT and COCO-LT datasets. Extensive ablation studies are conducted to verify the effectiveness of the essential proposals. © 1999-2012 IEEE.

关键词： Labeled data

来源：评论

学校读者我要写书评

暂无评论

A New Model for Network Security Situation Assessment of the Industrial Internet

引用

Computers, Materials & Continua 2023年第5期75卷 2527-2555页

作者： Ming Cheng Shiming Li Yuhe Wang Guohui Zhou Peng Han Yan Zhao College of Computer Science and Information Engineering Harbin Normal UniversityHarbin150025China School of Information Technology Luoyang Normal UniversityLuoyang471934China Shanghai Key Laboratory of Integrated Administration Technologies for Information Security Shanghai200240China

To address the problem of network security situation assessment in the Industrial Internet,this paper adopts the evidential reasoning(ER)algorithm and belief rule base(BRB)method to establish an assessment ***,this paper analyzes the influencing factors of the Industrial Internet and selects evaluation indicators that contain not only quantitative data but also qualitative ***,the evaluation indicators are fused with expert knowledge and the ER *** to the fusion results,a network security situation assessment model of the Industrial Internet based on the ER and BRB method is established,and the projection covariance matrix adaptive evolution strategy(P-CMA-ES)is used to optimize the model *** method can not only utilize semiquantitative information effectively but also use more uncertain information and prevent the problem of combinatorial ***,it solves the problem of the uncertainty of expert knowledge and overcomes the problem of low modeling accuracy caused by insufficient ***,a network security situation assessment case of the Industrial Internet is analyzed to verify the effectiveness and superiority of the *** research results showthat this method has strong applicability to the network security situation assessment of complex Industrial Internet *** can accurately reflect the actual network security situation of Industrial Internet systems and provide safe and reliable suggestions for network administrators to take timely countermeasures,thereby improving the risk monitoring and emergency response capabilities of the Industrial Internet.

关键词： Industrial internet network security situation assessment evidential reasoning belief rule base projection covariance matrix adaptive evolution strategy

来源：评论

学校读者我要写书评

暂无评论

Adaptive Filtering for UAV Integrated Navigation Based on Noise Variance

Adaptive Filtering for UAV Integrated Navigation Based on No...

引用

2023 China Automation Congress, CAC 2023

作者： Zhu, Jialu Zhao, Zhongyuan Ge, Quanbo Luo, Zijuan College of Automation Nanjing University of Information Science and Technology Nanjing China Technology GroupCorporation Key Laboratory of Information Systems Engineering The 28th Research Institute of ChinaElectronics Nanjing China

ISBN: (纸本)9798350303759

The Kalman filter is often used as the data fusion algorithm in the GPS/SINS integrated navigation system of the UAV. However, in practical applications, the traditional Kalman filter lacks the ability to deal with the statistical characteristics of noise, which are difficult to obtain and vary with the environment. Therefore, this paper first designs a noise estimator based on reliability for the problem that it is difficult to obtain the statistical characteristics of noise in practical applications;secondly, the measurement noise is updated separately after the Kalman measurement update to reduce the noise caused by noise changes. Finally, the simulation experiment verifies that the adaptive filter designed in this paper can achieve better data fusion in the application of UAV GPS/SINS integrated navigation, and has higher robustness and fault tolerance. © 2023 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Mn single-atoms decorated CNT electrodes for high-performance supercapacitors

引用

Nano Research 2024年第5期17卷 4039-4046页

作者： Qian Rong Chao Yuwen Yingkai Liu Yuanrong Liu Chunyang Wang Yang Wang Ding Zhang Zhen Wen Yunnan Key Laboratory of Opto-Electronic Information Technology Institute of Physics and Electronic InformationYunnan Normal UniversityKunming 650500China Faculty of Metallurgical and Energy Engineering Kunming University of Science and TechnologyKunming 650093China Department of Physics and Astronomy University of CaliforniaIrvineCA 92697USA School of Materials Science and Engineering National Institute for Advanced MaterialsSmart Sensing Interdisciplinary Science CenterNankai UniversityTianjin 300350China Jiangsu Key Laboratory for Carbon-Based Functional Materials and Devices Institute of Functional Nano and Soft Materials(FUNSOM)Soochow UniversitySuzhou 215123China

Developing highly robust and efficient electrode materials is of critical importance to promoting the energy density of current supercapacitors for ***,we report an efficient catalyst with monodispersed Mn single-atoms embedded in carbon nanotubes(Mn-CNTs)for enhancing the electrode performance of supercapacitors.A high specific capacitance(1523.6 F·g^(-1) at 1.0 A·g^(-1))can be achieved,which is about twice as high as the specific capacitance of the electrode material without the introduction of Mn ***,the asymmetric electrochemical capacitor created with Mn-CNT and activated carbon exhibits a high energy density of 180.8 Wh·kg^(-1) at a power density of 1.4 kW·kg^(-1),much higher than most reported *** study shows that the integration of Mn atoms into the CNT can enhance the charge transport capacity and the number of polar active sites of Mn-CNT and then facilitate chemical interactions between Mn-CNT and OH-.This work provides a novel strategy to enable high-energy storage in supercapacitors by introducing single-atoms into carbon nanotubes to improve electrodes’energy density and cycle life.

关键词： carbon nanotube(CNT) supercapacitor single-atoms power density energy density

来源：评论

学校读者我要写书评

暂无评论

A robust statistical prediction model for late-summer heavy precipitation days in North China

引用

science China Earth sciences 2025年第1期68卷 158-171页

作者： Shunli JIANG Tingting HAN Xin ZHOU Huijun WANG Zhicong YIN Xiaolei SONG Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/Key Laboratory of Meteorological Disaster Ministry of EducationNanjing University for Information Science and TechnologyNanjing210044China Nansen-Zhu International Research Centre Institute of Atmospheric PhysicsChinese Academy of SciencesBeijing100029China Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai) Zhuhai519080China

Recently,heavy precipitation(HP)events have occurred frequently in North China(NC),causing devastating economic losses and human ***,the short-term climate prediction of HP is quite *** year-to-year increment(DY)method and sliding correlations,we developed a robust seasonal prediction model for late-summer HP days(HPDs)in NC during 1982–2022,utilizing three independent predictors—February sea surface temperature(SST)in the Indian Ocean(SST_IO),February snow depth over North Asia(SDE_NA),and May melted snow depth in NC(MSDE_NC).The SST_IO anomalies affect NC's precipitation through the Pacific-Japan *** SDE_NA anomalies are associated with East Asian anomalous anticyclone by southeastern propagation of Rossby wave at *** MSDE_NC anomalies are followed by vertical motion and moisture anomalies in situ and thereby cause precipitation *** prediction model can well simulate the variations of the HPDs,with a correlation coefficient(CC)of 0.81(0.65)between the observed and predicted HPDs_DY(HPDs_anomaly).The percentage with the same sign for 15 extreme HPDs_anomaly years(PSSE)is 100%.Moreover,in the cross-validation test during 1982–2022,the PSSE for HPDs_anomaly is as high as 100%,along with a low rootmean-square error of *** independent hindcasts during 2013–2022,the CC between the observed and predicted HPDs_DY(HPDs_anomaly)is 0.93(0.83),together with high Nash-Sutcliffe efficiency(0.82)and agreement index(0.89).Specifically,the predictions are broadly consistent with the observations for 2015,2016,2017,2021,and 2022,reflecting excellent performance of this prediction model of HPDs in NC.

关键词： Heavy precipitation at North China Year-to-year increment approach Robust seasonal prediction Sea surface temperature Snow depth

来源：评论

学校读者我要写书评

暂无评论

FedIERF: Federated Incremental Extremely Random Forest for Wearable Health Monitoring

引用

Journal of Computer science & technology 2023年第5期38卷 970-984页

作者：胡春雨忽丽莎袁林陆佃杰吕蕾陈益强 Key Laboratory of Computing Power Network and Information Security Ministry of EducationShandong Computer Science Center(National Supercomputer Center in Jinan)Qilu University of Technology(Shandong Academy of Sciences)Jinan 250353China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer ScienceJinan 250000China Shandong Provincial Key Laboratory for Novel Distributed Computer Software Technology Jinan 25000China Institute of Information Technology Hebei University of Economics and BusinessShijiazhuang 050061China School of Information Science and Engineering Shandong Normal UniversityJinan 250358China Institute of Computing Technology Chinese Academy of SciencesBeijing 100190China

Wearable health monitoring is a crucial technical tool that offers early warning for chronic diseases due to its superior portability and low power ***,most wearable health data is distributed across dfferent organizations,such as hospitals,research institutes,and companies,and can only be accessed by the owners of the data in compliance with data privacy *** first challenge addressed in this paper is communicating in a privacy-preserving manner among different *** second technical challenge is handling the dynamic expansion of the federation without model *** address the first challenge,we propose a horizontal federated learning method called Federated Extremely Random Forest(FedERF).Its contribution-based splitting score computing mechanism significantly mitigates the impact of privacy protection constraints on model *** on FedERF,we present a federated incremental learning method called Federated Incremental Extremely Random Forest(FedIERF)to address the second technical *** introduces a hardness-driven weighting mechanism and an importance-based updating scheme to update the existing federated model *** experiments show that FedERF achieves comparable performance with non-federated methods,and FedIERF effectively addresses the dynamic expansion of the *** opens up opportunities for cooperation between different organizations in wearable health monitoring.

关键词： federated learning incremental learning random forest wearable health monitoring

来源：评论

学校读者我要写书评

暂无评论

Semantic Coherence Guided Multiview Similarity for Image Classification with Varied Supervision

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第6期5卷 3072-3082页

作者： Zhang, Chunjie Cao, Ping Bai, Huihui Zheng, Xiaolong Beijing Jiaotong University Institute of Information Science Beijing100044 China Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing100044 China University of Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems The State of Key Laboratory of Management and Control for Complex System Institute of Automation Beijing100190 China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing100190 China

Pairwise similarity has been widely used for image classification by propagating the class information from labeled images to unlabeled images and predicting the classes of unlabeled images accordingly. Although widely used, pairwise similarity based classification methods have two drawbacks. On one hand, visual-semantic similarity consistency does not always hold due to semantic gap. On the other hand, the reliability of pairwise similarity is also influenced by the number and underlying distribution of labeled images. A well-designed classification model cannot be easily generalized to few-shot classification tasks. To solve these two problems, in this article, we propose a semantic coherence guided multiview similarity (SCG-MV) for image classification method with varied supervision levels. We measure the semantic coherence between image pairs for classifier learning. This is achieved by first conducting line search along the line segment between each image pair, and then measure the semantics of each interval. The similarity of one image pair can then be obtained by aggregating the semantic coherence of all the intervals. In this way, we can model image similarities more finely and discriminatively. To make use of multiview information, we also use the semantic coherence guided image similarities to learn multiview classifiers. The proposed semantic coherence guided method can be combined with other similarity based methods for image classification with various levels of supervision. We conduct both few-shot and fully supervised classification experiments on four datasets with the results well prove the discriminative power and generalization ability of the proposed method. © 2020 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

A self-healing aqueous ammonium-ion micro batteries based on PVA-NH_(4)Cl hydrogel electrolyte and MXene-integrated perylene anode

引用

Nano Research Energy 2024年第4期3卷 57-67页

作者： Ke Niu Junjie Shi Long Zhang Yang Yue Mengjie Wang Qixiang Zhang Yanan Ma Shuyi Mo Shaofei Li Wenbiao Li Li Wen Yixin Hou Fei Long Yihua Gao College of Materials Science and Engineering Guangxi Key Laboratory of Optical and Electronic Materials and DevicesGuilin University of TechnologyGuilin 541004China School of Physics and Center for Nanoscale Characterization&Devices(CNCD) Wuhan National Laboratory for Optoelectronics(WNLO)Huazhong University of Science and Technology(HUST)Wuhan 430074China Information Materials and Intelligent Sensing Laboratory of Anhui Province Key Laboratory of Structure and Functional Regulation of Hybrid Materials of Ministry of EducationInstitutes of Physical Science and Information TechnologyAnhui UniversityHefei 230601China Hubei Key Laboratory of Critical Materials of New Energy Vehicles&School of Mathematics Physics and Optoelectronic Engineering Hubei University of Automotive TechnologyShiyan 442002China

The successful study of self-healing aqueous micro batteries(AMBs)which inherit the advantages of aqueous batteries and have the ability to automatically repair damage is of great significance for the development of smart wearable and portable electronic ***,the rate performance and the related power density of developed self-healing AMBs using metal ions as charge carriers is limited,due to the strong interaction between metal ions and electrode ***,there is great potential for developing self-healing NH^(4+)AMBs,because of the outstanding advantages of NH^(4+)such as extremely abundant reserves,smaller hydrated ion radius and little molar ***,the development of self-healing NH^(4+)AMBs is still an extremely challenge due to the difficulty in developing self-healing hydrogels and instability of anode *** though,the firstly self-healing NH^(4+)AMBs based on tailoring hydrogel electrolyte and MXene-integrated perylene anode were successfully *** expected,self-healing NH^(4+)AMBs exhibit excellent energy density(82.48µWh·cm^(-2))and power density(3.09 mW·cm^(-2)),cycle life(81.67%after 3000 GCD cycles),flexibility(95.68%under 180°)and self-healing ability(94.16%after the 10th self-healing cycles).

关键词： self-healing PVA-NH_(4)Cl hydrogel PTCDI-Ti_(3)C_(2)T_(x)MXene anode self-healing NH_(4)+AMBs high-rate performance flexibility

来源：评论

学校读者我要写书评

暂无评论

Polarization-directed nanophotonic routers based on twodimensional inorganic molecular crystals

引用

InfoMat 2024年第8期6卷 111-118页

作者： Jiacheng Yao Xin Feng Tingting Zhang Fangqi Chen Zhenglong Zhang Hairong Zheng Tianyou Zhai Tao Ding Key Laboratory of Artificial Micro/Nano Structure of Ministry of Education School of Physics and TechnologyWuhan UniversityWuhanthe People's Republic of China State Key Laboratory of Materials Processing and Die&Mould Technology and School of Materials Science and EngineeringHuazhong University of Science and TechnologyWuhanthe People's Republic of China Department of Mechanical Engineering The University of Texas at DallasRichardsonTexasUSA School of Physics and Information Technology Shaanxi Normal UniversityXi'anthe People's Republic of China

Photonic and plasmonic hybrid nanostructures are the key solution for integratednanophotonic circuits with ultracompact size but relative low loss. However,the poor tunability and modulability of conventional waveguides makesthem cumbersome for optical multiplexing. Here we make use of twodimensionalmolecular crystal, α-Sb_(2)O_(3) as a dielectric waveguide via totalinternal reflection, which shows polarization-sensitive modulation of the propagatingbeams due to its large polarization mode dispersion. Both experimentsand simulations are performed to verify such concept. These Sb_(2)O_(3) nanoflakescan be coupled with plasmonic nanowires to form nanophotonic beam splittersand routers which can be easily modulated by changing the polarization of theincidence. It thus provides a robust, exploitable and tunable platform for onchipnanophotonics.

关键词： nanoflakes plasmonics polarization Sb_(2)O_(3) waveguides

来源：评论

学校读者我要写书评

暂无评论

Geometric error measuring,modeling,and compensation for CNC machine tools:A review

引用

Chinese Journal of Aeronautics 2024年第2期37卷 163-198页

作者： Zhao ZHANG Feng JIANG Ming LUO Baohai WU Dinghua ZHANG Kai TANG Engineering Research Center of Advanced Manufacturing Technology for Aero Engine(Northwestern Polytechnical University) Ministry of EducationXi’an 710072China Key Laboratory of High Performance Manufacturing for Aero Engine(Northwestern Polytechnical University) Ministry of Industry and Information TechnologyXi’an 710072China Department of Mechanical and Aerospace Engineering Hong Kong University of Science and TechnologyHong KongChina

Geometric error,mainly due to imperfect geometry and dimensions of machine components,is one of the major error sources of machine *** that geometric error has significant effects on the machining quality of manufactured parts,it has been a popular topic for academic and industrial research for many years.A great deal of research work has been carried out since the 1970s for solving the problem and improving the machining *** have studied how to measure,detect,model,identify,reduce,and compensate the geometric *** paper presents a thorough review of the latest research activities and gives an overview of the state of the art in understanding changes in machine tool performance due to geometric *** advances in measuring the geometrical errors of machine tools are summarized,and different kinds of error identification methods of translational axes and rotation axes are illustrated ***,volumetric geometric error modeling,tracing,and compensation techniques for five-axis machine tools are emphatically ***,research challenges in order to improve the volumetric accuracy of machine tools are also highlighted.

关键词： Error compensation Error identification Error measurement Error modeling Geometric error Machine tools

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：