检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Shibuya, Tatsukichi Inoue, Nakamasa Kawakami, Rei Sato, Ikuro Department of Computer Science Institute of Science Tokyo Tokyo Japan Department of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan Denso IT Laboratory Tokyo Japan

ISBN: (纸本)9798350368741

For deploying deep neural networks on edge devices with limited resources, binary neural networks (BNNs) have attracted significant attention, due to their computational and memory efficiency. However, once a neural network is binarized, finetuning it on edge devices becomes challenging because most conventional training algorithms for BNNs are designed for use on centralized servers and require storing real-valued parameters during training. To address this limitation, this paper introduces binary stochastic flip optimization (BinSFO), a novel training algorithm for BNNs. BinSFO employs a parameter update rule based on Boolean operations, eliminating the need to store real-valued parameters and thereby reducing memory requirements and computational overhead. In experiments, we demonstrated the effectiveness and memory efficiency of BinSFO in finetuning scenarios on six image classification datasets. BinSFO performed comparably to conventional training algorithms with a 70.7% smaller memory requirement. Code is released at https://***/TatsukichiShibuya/ICASSP2025_BinSFO. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Real-Time Semantic Segmentation Method Based on Transformer for Autonomous Driving

引用

computers, Materials & Continua 2024年第12期81卷 4419-4433页

作者： Weiyu Hao Jingyi Wang Huimin Lu Donald Bren School of Information and Computer Sciences University of CaliforniaIrvineCA 92612USA Department of Control Engineering Kyushu Institute of TechnologyKitakyushu804-8550Japan School of Automation Southeast UniversityNanjing210096China

While traditional Convolutional Neural Network(CNN)-based semantic segmentation methods have proven effective,they often encounter significant computational challenges due to the requirement for dense pixel-level predictions,which complicates real-time *** address this,we introduce an advanced real-time semantic segmentation strategy specifically designed for autonomous driving,utilizing the capabilities of Visual *** leveraging the self-attention mechanism inherent in Visual Transformers,our method enhances global contextual awareness,refining the representation of each pixel in relation to the overall *** enhancement is critical for quickly and accurately interpreting the complex elements within driving sce-narios—a fundamental need for autonomous *** experiments conducted on the DriveSeg autonomous driving dataset indicate that our model surpasses traditional segmentation methods,achieving a significant 4.5%improvement in Mean Intersection over Union(mIoU)while maintaining real-time *** paper not only underscores the potential for optimized semantic segmentation but also establishes a promising direction for real-time processing in autonomous navigation *** work will focus on integrating this technique with other perception modules in autonomous driving to further improve the robustness and efficiency of self-driving perception frameworks,thereby opening new pathways for research and practical applications in scenarios requiring rapid and precise decision-making *** experimentation and adaptation of this model could lead to broader implications for the fields of machine learning and computer vision,particularly in enhancing the interaction between automated systems and their dynamic environments.

关键词： Visual transformer semantic segmentation autonomous driving

来源：评论

学校读者我要写书评

暂无评论

AFSTGCN:Prediction for multivariate time series using an adaptive fused spatial-temporal graph convolutional network

引用

Digital Communications and Networks 2024年第2期10卷 292-303页

作者： Yuteng Xiao Kaijian Xia Hongsheng Yin Yu-Dong Zhang Zhenjiang Qian Zhaoyang Liu Yuehan Liang Xiaodan Li School of Information and Control Engineering China University of Mining and TechnologyXuzhou221116China Department of Informatics University of LeicesterLeicesterLE17RHUK The Affiliated Changshu Hospital of Soochow University Changshu215500China School of Computer Science and Engineering Changshu Institute of TechnologySuzhou215500China

The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other ***,it is important to construct a digital twin ***,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted *** this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)***,we fuse the spatial-temporal graph based on the interrelationship of spatial ***,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding ***,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)*** module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like *** dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted *** on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.

关键词： Adaptive adjacency matrix Digital twin Graph convolutional network Multivariate time series prediction Spatial-temporal graph

来源：评论

学校读者我要写书评

暂无评论

Biomedically Informed ECG Synthesis: Customizing Cardiac Cycle Phases with Diffusion Model

Biomedically Informed ECG Synthesis: Customizing Cardiac Cyc...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Lin, Yuxin Ma, Jing Wang, Wei Wu, Zhihao Dong, Suyu Luo, Gongning Wang, Kuanquan Harbin Institute of Technology School of Computer Science and Technology Shenzhen China Northeast Forestry University College of Computer and Control Engineering Harbin China Harbin Institute of Technology Faculty of Computing Harbin China

ISBN: (纸本)9798350386226

Cardiovascular diseases are a major global health challenge, with electrocardiography (ECG) being critical for diagnosis and monitoring. As artificial intelligence and automated ECG diagnostic technologies rapidly advance, the demand for large-scale ECG databases continues to grow. Generative ECG has become a mainstream method to enhance database size and diversity. However, existing methods typically generate ECG randomly or focus on limited physiological categories, lacking the ability to synthesize ECG with varying physiological features and cardiac cycles, which is crucial for various practical applications. In response to this need, we propose a novel approach introducing a diffusion model called DIFF-ECG to generate precisely customized ECG that accurately reflect diverse cardiac conditions. Segmentation-based quality assessments confirmed that the synthesized ECG accurately followed the specified cardiac cycle information, with our model significantly outperforming baseline diffusion and GAN-based methods. Therefore, our approach addresses the critical need for generating clinically relevant and customizable ECG, contributing significantly to the field of automated cardiac disease diagnosis. By enabling fine-tuning of cardiac cycle phases, our method significantly expands the application range of generative ECG, potentially improving the diagnostic accuracy for rare diseases and advancing personalized medicine. © 2024 IEEE.

关键词： Electrocardiography

来源：评论

学校读者我要写书评

暂无评论

Path Planning and control of a 6-DOF Manipulator for Freeform Surface Polishing Based-on a Point Cloud Model

Path Planning and Control of a 6-DOF Manipulator for Freefor...

引用

2024 International Conference on Advanced Robotics and Intelligent Systems, ARIS 2024

作者： Tsai, Fu-Yu Song, Kai-Tai Institute of Electrical and Control Engineering College of Electrical and Computer Engineering National Yang Ming Chiao Tung University 1001 Ta Hsueh Road Hsinchu Taiwan

ISBN: (纸本)9798350362572

This paper presents a novel path generation method for robotic surface polishing based on point cloud data. In this method, an initial path is first planned to create waypoints via the point cloud. The method estimates the contact area to dynamically adjust initial path points to ensure uniform trajectory coverage on the surface. In this paper, the bilateral tool path generation (BTPG) algorithm is proposed based on a dual-thread strategy to expedite polishing path generation. Further, we introduce a method that generates smooth postures to enhance the stability of the robotic polishing pose. In this work, hybrid force/velocity control scheme is adopted to track polishing path and apply the assigned force on the workpiece freeform surface. Practical experimental results show that the proposed method achieves 95% improvement in path planning time and 85% in smoothness of the robot fourth joint's posture compared with the baseline slicing method. © 2024 IEEE.

关键词： Robot programming

来源：评论

学校读者我要写书评

暂无评论

MPC-based Optimization Design for 3D Collision Avoidance of a Mobile Manipulator Based-on Obstacle Velocity Estimation

MPC-based Optimization Design for 3D Collision Avoidance of ...

引用

2024 International Automatic control Conference, CACS 2024

作者： Song, Kai-Tai Lin, Chih-Hsuan Institute of Electrical and Control Engineering College of Electrical and Computer Engineering National Yang Ming Chiao Tung University 1001 Ta Hsueh Road Hsinchu Taiwan

ISBN: (纸本)9798350354904

This paper presents an optimization design of 3D dynamic obstacle avoidance for a mobile manipulator based on model predictive control (MPC). The design enables a mobile manipulator to achieve optimized 3D collision avoidance motion with shorter avoidance path and faster avoidance time. A 3D LiDAR is installed onboard the robot to acquire environmental point cloud and estimate obstacle velocity. The MPC is designed to track an initial 3D path of the mobile manipulator and avoid any static and dynamic obstacles in real time. Experimental results show that the proposed method can simultaneously avoid static and dynamic obstacles in 3D space. Compared with baseline algorithms without velocity estimation, the proposed method reduces the avoidance path length by 8.27% and path execution time by 13.79%. © 2024 IEEE.

关键词： Predictive control systems

来源：评论

学校读者我要写书评

暂无评论

Comparison of Two AC Current De-tumbling Methods for Uncooperative Space Target through Electromagnetic Characteristic Analysis 1

Comparison of Two AC Current De-tumbling Methods for Uncoope...

引用

1st Aerospace Frontiers Conference, AFC 2024

作者： Du, Lei Chen, Zhen Hu, Hengzai Liu, Xiangdong Zhang, Qiang Zhang, Kemo Liu, Hui School of Automation Beijing Institute of Technology Beijing100081 China Beijing Institute of Control Engineering Beijing100191 China Beijing Institute Computer for Application Beijing100089 China

ISBN: (纸本)9781510681613

A large number of space debris with rotational motion occupy scarce orbital resources, thus, a de-tumbling phase may be necessary prior to the capturing phase to reduce the risk of collision. A new contactless AC current de-tumbling method based on the principle of electromagnetic eddy current is proposed in this paper. Firstly, an analysis is conducted on external disturbances affecting the attitude of the debris and de-tumbling device, such as magnetic field perturbations. Compared to the electromagnetic forces acting on de-tumbling device, the influence of the geomagnetic force on detumbling device can be negligible. However, the geomagnetic torque acting on de-tumbling device cannot be ignored. Then, the advantages and disadvantages of the two AC current cases are compared and analyzed, and the method to increase the de-tumbling torque can be obtained by adjusting the frequency of the three phase AC current. Finally, the three-dimensional attitude dynamics of the uncooperative space target is established taking a spinning cylinder as the hypothetical target of de-tumbling. Simulation shows that the de-tumbling method is effective and the de-tumbling rate can be greatly improved by adjusting the frequency of the three phase AC current. © 2024 SPIE.

关键词： Space debris

来源：评论

学校读者我要写书评

暂无评论

A novel fractional physics-informed neural networks method for solving the time-fractional Huxley equation

引用

Neural Computing and Applications 2024年第30期36卷 19097-19119页

作者： Shi, Jieyu Yang, Xiaozhong Liu, Xinlong School of Control and Computer Engineering North China Electric Power University Beijing102206 China Institute of Information and Computing School of Mathematics and Physics North China Electric Power University Beijing102206 China

The neural network methods in solving differential equations have significant research importance and promising application prospects. Aimed at the time-fractional Huxley (TFH) equation, we propose a novel fractional physics-informed neural networks (fPINNs) method. By integrating the physical information of the TFH equation into neural networks, the fPINNs are trained as a precise approximation model for solving the TFH equation. The fPINNs method involves calculating the Caputo fractional derivative using the L1 formula and estimating the integer-order derivative through the chain rule. The Adam algorithm is employed to optimize two kinds of loss functions constructed by hard and soft constraints, respectively. Through an analysis of the impact of fPINNs parameters on training results, we identify the optimal parameter configuration for solving both one-dimensional and two-dimensional (2D) TFH equations. Numerical examples validate the efficiency and robustness of the fPINNs method in solving TFH equations. Furthermore, numerical simulation of the 2D TFH problem demonstrates the method’s practical applicability to engineering problem. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

Point Cloud Classification Using Content-Based Transformer via Clustering in Feature Space

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 231-239页

作者： Yahui Liu Bin Tian Yisheng Lv Lingxi Li Fei-Yue Wang the State Key Laboratory for Management and Control of Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190 the School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100190 IEEE the Transportation and Autonomous Systems Institute(TASI)and the Department of Electrical and Computer Engineering Purdue School of Engineering and TechnologyIndiana University-Purdue University Indianapolis(IUPUI)Indianapolis 46202 USA

Recently, there have been some attempts of Transformer in 3D point cloud classification. In order to reduce computations, most existing methods focus on local spatial attention,but ignore their content and fail to establish relationships between distant but relevant points. To overcome the limitation of local spatial attention, we propose a point content-based Transformer architecture, called PointConT for short. It exploits the locality of points in the feature space(content-based), which clusters the sampled points with similar features into the same class and computes the self-attention within each class, thus enabling an effective trade-off between capturing long-range dependencies and computational complexity. We further introduce an inception feature aggregator for point cloud classification, which uses parallel structures to aggregate high-frequency and low-frequency information in each branch separately. Extensive experiments show that our PointConT model achieves a remarkable performance on point cloud shape classification. Especially, our method exhibits 90.3% Top-1 accuracy on the hardest setting of ScanObjectN N. Source code of this paper is available at https://***/yahuiliu99/PointC onT.

关键词： Content-based Transformer deep learning feature aggregator local attention point cloud classification

来源：评论

学校读者我要写书评

暂无评论

Multi-scale network for single image deblurring based on ensemble learning module

引用

Multimedia Tools and Applications 2025年第11期84卷 9045-9064页

作者： Wu, WenBo Pan, Yun Su, Na Wang, JingTao Wu, ShaoChuan Xu, ZeKun Yu, YouJian Liu, YaPeng State Key Laboratory of Media Convergence and Communication Communication University of China Beijing100024 China School of Computer and Cyberspace Security Communication University of China Beijing100024 China School of Electronics and Information Engineering Harbin Institute of Technology Harbin150080 China College of Computer and Control Engineering Northeast Forestry University Harbin150040 China School of Computer and Information Engineering Tianjin Chengjian University Tianjin300384 China School of Computer Science and Technology Anhui University Hefei230601 China

In this paper, we have identified two primary issues with current multi-scale image deblurring methods. On the one hand, the blurring scale is ignored. On the other hand, the context information of images is not fully utilized, and the spatial detail information of images is lost. To better solve the above problems, on top of that, this paper proposes a multi-scale network for single image deblurring. The designed network contains two parts: 1) Ensemble Learning Model: The Weak Learning Model in this module connects multiple weak learners in series to increase network depth. Moreover, through multi-scale feature ensemble module, the output features and intermediate features of different weak learner are integrated, and finally strong learner is used to enhance the features with semantic information at different scales;2) Spatial Feature Enhancement Module: This module can preserve spatial detail information of images and compensate for the disadvantages of losing image spatial details due to continuous downsampling in the Ensemble Learning Model structure. Our proposed deblurring model has been extensively evaluated on several benchmark datasets, and the model achieves superior performance compared to the most advanced deblurring methods. In particular, the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) values are 33.57 dB and 0.965 on the GoPro dataset, respectively, with PSNR and SSIM improving by 0.49 dB and 0.002 over NAFNet, and 0.65 dB and 0.004 over Restormer. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Learning systems

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：