检索结果-内蒙古大学图书馆

Predicting Bird's-Eye-View Semantic Representations Using Correlated Context Learning

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第5期9卷 4718-4725页

作者： Chen, Yongquan Fan, Weiming Zheng, Wenli Huang, Rui Yu, Jiahui Chinese Univ Hong Kong Shenzhen Inst Artificial Intelligence & Robot Soc Shenzhen 518172 Peoples R China Shenyang Ligong Univ Sch Automat & Elect Engn Shenyang 110159 Peoples R China Dapeng Customs Peoples Republ China Shenzhen 518083 Peoples R China Zhejiang Univ Dept Biomed Engn Hangzhou 310027 Peoples R China

We redefine the concept of bird's-eye-view (BEV) imaging for machine cognition tasks, emphasizing its power as an image interpretation tool. Humans intuitively translate two-dimensional (2D) images into BEV representations by discerning and integrating spatial information, such as position and morphological aspects. Existing techniques focus primarily on improving accuracy in whole-to-whole mapping. However, this often results in a loss of global-local correlation, posing a significant challenge in predicting complex elements, such as multiscale dynamic objects and small-scale static objects in the distance. To address this issue, we propose correlated global-local spatial context learning (CGLSCL), one of the first attempts to amalgamate positional and morphological cues in translation for machine cognition tasks. Augmented by correlated learning, CGLSCL ensures more comprehensive BEV output, particularly for minor and fast-moving elements, which need to be captured more effectively than they are by existing methods. An evaluation of CGLSCL using the NuScenes and Argoverse 3D datasets demonstrated its superior performance compared to current state-of-the-art methods, particularly in predicting complex elements.

关键词： BEV machine cognition attention transformers

来源：评论

学校读者我要写书评

暂无评论

Cross-User Leakage Mitigation for Authorized Multi-User Encrypted Data Sharing

引用

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 2024年 19卷 1213-1226页

作者： Wang, Mingyue Chen, Zizhuo Miao, Yinbin Huang, Hejiao Wang, Cong Jia, Xiaohua Harbin Inst Technol Sch Comp Sci & Technol Shenzhen 518055 Peoples R China City Univ Hong Kong Dept Comp Sci Hong Kong Peoples R China Xidian Univ Sch Cyber Engn Xian 710071 Peoples R China City Univ Hong Kong Shenzhen Res Inst Shenzhen 518057 Peoples R China

Cloud computing has been a research focus in both academic and industrial communities for decades. Along with this trend, Searchable Encryption (SE) technology emerged and developed as data privacy concerns increased. Many schemes are proposed to solve the privacy-preserving data-sharing problem in multi-user scenarios. Most existing solutions are based on the assumption that all users are trusted. However, there will be cross-user leakage when there are malicious or compromised ones. This is because of the inherent linkability of authorization information and the search result when multiple users request data from the same database. To this end, we propose a cross-user leakage mitigation scheme for authorized encrypted data sharing in a two-server model. We utilize a blinding factor to delink authorizations based on Symmetric Multi-Key Searchable Encryption (SMKSE). To break the linkability of query results, we combine the zero-sum garbled Bloom filter with the oblivious transfer technique, where each of the two servers can only know partial information. We devise a group-based Bloom filter structure in indices to improve efficiency. We perform formal security analysis and also demonstrate the efficiency through comparative experiments.

关键词： Searchable encryption cross-user leakage garbled Bloom filter oblivious transfer

来源：评论

学校读者我要写书评

暂无评论

Deep Source Semi-Supervised Transfer Learning (DS3TL) for Cross-Subject EEG Classification

引用

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 2024年第4期71卷 1308-1318页

作者： Jiang, Xue Meng, Lubin Wang, Ziwei Wu, Dongrui Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Minist Educ Image Proc & Intelligent Contr Wuhan 430074 Peoples R China Shenzhen Huazhong Univ Sci & Technol Res Inst Shenzhen 518000 Peoples R China

Objective: An electroencephalogram (EEG) based brain-computer interface (BCI) maps the user's EEG signals into commands for external device control. Usually a large amount of labeled EEG trials are required to train a reliable EEG recognition model. However, acquiring labeled EEG data is time-consuming and user-unfriendly. Semi-supervised learning (SSL) and transfer learning can be used to exploit the unlabeled data and the auxiliary data, respectively, to reduce the amount of labeled data for a new subject. Methods: This paper proposes deep source semi-supervised transfer learning (DS3TL) for EEG-based BCIs, which assumes the source subject has a small number of labeled EEG trials and a large number of unlabeled ones, whereas all EEG trials from the target subject are unlabeled. DS3TL mainly includes a hybrid SSL module, a weakly-supervised contrastive module, and a domain adaptation module. The hybrid SSL module integrates pseudo-labeling and consistency regularization for SSL. The weakly-supervised contrastive module performs contrastive learning by using the true labels of the labeled data and the pseudo-labels of the unlabeled data. The domain adaptation module reduces the individual differences by uncertainty reduction. Results: Experiments on three EEG datasets from different tasks demonstrated that DS3TL outperformed a supervised learning baseline with many more labeled training data, and multiple state-of-the-art SSL approaches with the same number of labeled data. Significance: To our knowledge, this is the first approach in EEG-based BCIs that exploits the unlabeled source data for more accurate target classifier training.

关键词： Brain-computer interface electroencephalogram semi-supervised learning transfer learning classification

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Dispersion Spectrum Inversion for Surface Wave Exploration

引用

IEEE TRANSACTIONS ON GEOscience AND REMOTE SENSING 2024年 62卷 1页

作者： Gan, Yuandi Yang, Zhentao Pan, Lei Sun, Yao-Chong Zhang, Dazhou Gao, Yuqiu Chen, Xiaofei Southern Univ Sci & Technol Dept Earth & Space Sci Shenzhen 518055 Peoples R China Chengdu XWFintech Serv Co Ltd Chengdu 610000 Peoples R China Southern Univ Sci & Technol Guangdong Prov Key Lab Geophys High Resolut Imagin Shenzhen 518055 Peoples R China Tongji Univ Sch Ocean & Earth Sci State Key Lab Marine Geol Shanghai 200092 Peoples R China Cent South Univ Sch Geosci & Info Phys Changsha 410082 Peoples R China

The conventional surface wave inversion methodology relying upon dispersion curve analyses is frequently confounded by the phenomenon known as "mode kissing" as well as the absorptive properties of seismic waves traversing near-surface geologic strata. We propose a deep learning approach to overcome challenges in surface wave inversion due to complex near-surface geology. A convolutional neural network regression model is developed to directly map surface wave dispersion spectra to shear wave velocity profiles. The network takes entire spectrograms as input, avoiding issues from manual picking of individual dispersion curves. This circumvents problems of mode misidentification and crossover that hamper traditional inversion in layered media. Furthermore, we incorporate the quality factor Q into calculating the theoretical frequency dispersion spectrum to account for the absorptive influence of geologic formations. Through studying theoretical spectra, we found that when the Q-factor is less than 10, the absorption of the formation will have a greater impact on the dispersion imaging. Compared to existing workflows requiring manual parameter tuning and computationally expensive inversion calculations, the trained network directly infers velocities without user intervention or iterative optimization. The end-to-end relation learned from synthetic training data enables rapid interpretation of field records complicated by low-velocity zones and high-velocity inclusions, which inhibit conventional inversion algorithms.

关键词： Deep learning Rayleigh wave surface wave the kernel of Green's function (KGF)

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE-ACM TRANSACTIONS ON NETWORKING 2024年第3期32卷 2085-2098页

作者： Yao, Xiaopeng Zhao, Yunpeng Gao, Ningtuo Du, Hongwei Huang, Hejiao Harbin Inst Technol Shenzhen 518055 Guangdong Peoples R China Zhejiang Key Lab Social Secur Big Data Hangzhou 310000 Zhejiang Peoples R China

There is a huge amount of information generated in online social networks, which is filled with a lot of rumors. The spread of a rumor often leads to the generation of a causal related rumor, and when users believe the first kind of rumor, the probability of being influenced by another causal related rumor is larger. Therefore, the influence probability will change with the process of rumor spreading. In this paper, we design the Causal Rumors Enhance Cascade (CREC) model to describe the spreading process of causal related rumors. Then we attempt to select a set of seed users that minimizes the number of users expected to be influenced by rumors, which we call the Causal Related Rumors Controlling (CRRC) problem. The main challenges of this problem are that the influence probability is constantly changing during the spread process, so the reverse sampling technique cannot be used, and the greedy mechanism is not suitable for massive-scale datasets. For the sake of overcoming these challenges and solving the problem, we put forward the Degree Trigonometric Metrology (DTM) algorithm, which uses the property of three-directed circles in the directed network to select seed nodes. Finally, experiments on three massive-scale datasets show that our algorithm outperforms the other algorithms.

关键词： Causal related rumors social network massive-scale three-directed circle

来源：评论

学校读者我要写书评

暂无评论

A Less-Delay and Wide Conversion-Voltage-Range Level Shifter by Using Switched-Capacitor Technique for Power-Efficiency Health Electronics

引用

IEEE TRANSACTIONS ON CONSUMER ELECTRONICS 2024年第1期70卷 2382-2390页

作者： Duan, Quanzhen Tao, Tao Weng, Wuxiong Huang, Shengming Liang, Yuhua Roh, Jeongjin Xing, Xinpeng Sun Yat Sen Univ Shenzhen Sch Integrated Circuits Shenzhen 518107 Peoples R China Tianjin Univ Technol Sch Integrated Circuit Sci Engn Tianjin 300387 Peoples R China Xidian Univ Key Lab Analog Integrated Circuits & Syst Minist Educ Xian 710071 Peoples R China Xidian Univ Sch Microelect Xian 710071 Peoples R China Hanyang Univ Dept Elect Engn Ansan 426791 South Korea

This study proposes a novel less-delay and wide conversion-voltage-range (CVR) level shifter (LS) for high power-efficiency healthy electronics. The proposed LS consisting of a control signal generator with the switched-capacitor technique (CSG-SC), an internal converting stage (ICS), a latch-up circuit, and a LS core based on the differential cascade voltage switch (DCVS) was implemented by both high-threshold-voltage (HTV) and low-threshold-voltage (LTV) transistors. Switches and capacitors controlled by input-clock signals generating by CSG-SC are used to respectively shift up the voltages (at the top plate) of two capacitors and obtain higher-voltage control signals to turn on HTV transistor switches in ICS. The ICS working with the latch-up circuit obtained the level-shifted output signals, and finally a fast output clock signal with LS core based on DCVS is generated. The proposed LS has been implemented in a standard 180 nm CMOS process, and simulation results show that it achieves a large CVR from 0.6 to a 3 V, with a propagation delay as low as 2.5 ns, and a total average power of 704 nw for a 1 MHz input pulse.

关键词： Transistors Integrated circuits Switching circuits Switches Voltage control Delays Control systems Level shifter wide conversion-voltage range switched-capacitor health electronics

来源：评论

学校读者我要写书评

暂无评论

Design of Planar Magnetic Integrated LCL-EMI Hybrid Filter for the Grid-Connected Inverter

引用

IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS 2024年第3期60卷 4280-4291页

作者： Liu, Yitao Zhang, Xudong Lu, Zijian Yin, Jian Shenzhen Univ Coll Mechatron & Control Engn Shenzhen 518060 Peoples R China Shenzhen Univ Guangdong Key Lab Electromagnet Control & Intellig Shenzhen 518000 Peoples R China

Numerous harmonics and electromagnetic interference (EMI) noise are prevalent in inverter system, potentially compromising output power quality. Harmonic and EMI filters serve as crucial components for mitigating harmonic distortion and EMI noise in grid-connected inverter. Nonetheless, these filters significantly increase the overall weight and volume of the inverter system, so there is a need to reduce their size and weight. This article uses planar magnetic integration technology to integrate the inductors of the LCL filter, the inductors of the passive EMI filter and the sampling coil of the active EMI filter into a single magnetic core unit through reasonable design of the PCB windings. In addition, capacitance integration is facilitated by the insertion of dielectric material between planar coil layers. This innovative approach effectively reduces the size and weight of the filter, consequently elevating the power density of the inverter system. The planar magnetic integrated LCL-EMI hybrid filter is analyzed using a single-phase Gallium-Nitride (GaN) inverter simulation and experimental platform, demonstrating that the design scheme reduces the size and weight of the filter while ensuring harmonic and EMI suppression. The feasibility of the planar magnetic integration scheme is thereby verified.

关键词： Power harmonic filters Magnetic separation Electromagnetic interference Inductors Magnetic flux Windings Magnetic cores Active emi filter inverter LCL filter passive emi filter planar magnetic integration

来源：评论

学校读者我要写书评

暂无评论

CAT-DTI: cross-attention and Transformer network with domain adaptation for drug-target interaction prediction

引用

BMC BIOINFORMATICS 2024年第1期25卷 141页

作者： Zeng, Xiaoting Chen, Weilin Lei, Baiying Shenzhen Univ Sch Comp & Software Engn Shenzhen 518060 Peoples R China Shenzhen Univ Med Sch Marshall Lab Biomed Engn Shenzhen 518055 Peoples R China Shenzhen Univ Sch Biomed Engn Shenzhen 518055 Peoples R China

Accurate and efficient prediction of drug-target interaction (DTI) is critical to advance drug development and reduce the cost of drug discovery. Recently, the employment of deep learning methods has enhanced DTI prediction precision and efficacy, but it still encounters several challenges. The first challenge lies in the efficient learning of drug and protein feature representations alongside their interaction features to enhance DTI prediction. Another important challenge is to improve the generalization capability of the DTI model within real-world scenarios. To address these challenges, we propose CAT-DTI, a model based on cross-attention and Transformer, possessing domain adaptation capability. CAT-DTI effectively captures the drug-target interactions while adapting to out-of-distribution data. Specifically, we use a convolution neural network combined with a Transformer to encode the distance relationship between amino acids within protein sequences and employ a cross-attention module to capture the drug-target interaction features. Generalization to new DTI prediction scenarios is achieved by leveraging a conditional domain adversarial network, aligning DTI representations under diverse distributions. Experimental results within in-domain and cross-domain scenarios demonstrate that CAT-DTI model overall improves DTI prediction performance compared with previous methods.

关键词： Drug-target interaction Transformer Cross-attention Domain adaptation

来源：评论

学校读者我要写书评

暂无评论

Robot Learning From Demonstration for Assembly With Sequential Assembly Movement Primitives

引用

IEEE-ASME TRANSACTIONS ON MECHATRONICS 2024年第4期29卷 2685-2696页

作者： Hu, Haopeng Yan, Hengyuan Yang, Xiansheng Lou, Yunjiang Harbin Inst Technol Shenzhen Sch Mech Engn & Automat Guangdong Prov Key Lab ofIntelligent Morphing Mech Shenzhen Key Lab Adv Mot Control Technol & Modern Shenzhen 518055 Peoples R China

To facilitate the labor-consuming flexible assembly lines with robots, the robot learning from demonstration (LfD) is the promising way to efficiently impart human assembly skills to robots. Aiming at the challenging complex precise assembly tasks, which are contact-rich and require 6-D movement, an LfD method is proposed here. Due to the inconsistent requirements for the robot's assembly movement, the whole robotic assembly processes are composed of three phases, namely, the approaching, aligning, and assembling phase. The policies, which are prestructured by the proposed sequential assembly movement primitives, are learned exclusively to guide the robot's movement in each phase. In the approaching phase, the policy generates a reliable path for the robot to accurately track. However, in the aligning and assembling phase, the polices enable the robot's active compliant behavior to accomplish the complex precise assembly task. Robotic assembly experiments with four objects are conducted to validate the proposed LfD methods with a torque-controlled robot. Experiment results indicate that the proposed LfD method applied with the proposed policies achieves high reliability and efficiency.

关键词： Compliant control flexible manufacturing learning from demonstration (LfD) robotic assembly

来源：评论

学校读者我要写书评

暂无评论

Traction Network Wire Current Detection Based on Ring-Shaped TMR Array

引用

IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2024年第2期10卷 4268-4278页

作者： Xu, X. P. Yang, L. X. Chen, H. Y. Liu, T. Z. Xu, X. L. Sun, Z. Liu, H. L. Shanghai Publishing & Printing Coll Dept Informat & Intelligent Engn Shanghai 200093 Peoples R China Zhejiang Acad Special Equipment Sci Hangzhou 310009 Peoples R China Shanghai Univ Sch Mechatron Engn & Automat Shanghai 200044 Peoples R China Zhengzhou Univ Affiliated Hosp 1 Zhengzhou 450000 Peoples R China Chinese Univ Hong Kong Sch Sci & Engn Shenzhen 518172 Peoples R China Shenzhen Inst Artificial Intelligence & Robot Soc Shenzhen 518116 Peoples R China

Ensuring safe operation of electrified railways is critical, and one challenge is detecting the current in railway traction network cables to prevent them from burning down due to excessive current. This article analyzes the working principle of the ring-shaped tunnel magnetoresistance (TMR) array current sensor. To detect the current in the traction network load-bearing cable without interrupting or dismantling the transmission line, an open-close structure is adopted in the sensor design. Finite element simulation and experimental tests verify the rationality of the proposed scheme, showing that the current detection error is within 0.5%.

关键词： Sensitivity Rail transportation Current measurement Resistance Sensor arrays Conductors Bridge circuits Current sensor finite element simulation open-close structure railway traction network tunnel magnetoresistance (TMR)

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：