检索结果-内蒙古大学图书馆

第41届中国控制会议

作者： Mohan Jia Zhongjian Dai Yaping Dai Zhiyang Jia School of Automation Beijing Institute of Technology State Key Laboratory of Intelligent Control and Decision of Complex Systems

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

In one-stage methods for video moment retrieval,the common representations indirectly supervised by boundary prediction fail to fully preserve the inherent characteristic of the video and query,which limits the retrieval *** solve this problem,an Adversarial Video Moment Retrieval(AVMR) algorithm is proposed to learn the common representations with modality invariance and cross-modal *** is implemented through the process of adversarial learning between a feature projector and a modality *** feature projector tries to generate a modality-invariant common representation and to confuse the modality *** modality classifier tries to discriminate between different modalities based on the generated representation by the feature *** triplet constraints are further imposed on the feature projector to preserve the underlying cross-modal semantic structure of *** experimental results show that AVMR surpasses the baseline Attentive Cross-modal Relevance Matching(ACRM) by 1.10% and 1.73% in the "mIoU" metric on two public datasets Charades-STA and TACoS,respectively.

关键词： Video Moment Retrieval Cross-modal Retrieval Deep learning Adversarial Learning

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Padding Correlation Filter With Group Feature Fusion for Robust Visual Tracking

引用

IEEE/CAA Journal of Automatica Sinica 2022年第10期9卷 1845-1860页

作者： Zihang Feng Liping Yan Yuanqing Xia Bo Xiao Key Laboratory of Intelligent Control and Decision of Complex Systems the School of AutomationBeijing Institute of TechnologyBeijing 100081China IEEE the School of Artificial Intelligence Beijing University of Posts and TelecommunicationsBeijing 100876China

In recent visual tracking research,correlation filter(CF)based trackers become popular because of their high speed and considerable *** methods mainly work on the extension of features and the solution of the boundary effect to learn a better correlation ***,the related studies are *** exploring the potential of trackers in these two aspects,a novel adaptive padding correlation filter(APCF)with feature group fusion is proposed for robust visual tracking in this paper based on the popular context-aware tracking *** the tracker,three feature groups are fused by use of the weighted sum of the normalized response maps,to alleviate the risk of drift caused by the extreme change of single ***,to improve the adaptive ability of padding for the filter training of different object shapes,the best padding is selected from the preset pool according to tracking precision over the whole video,where tracking precision is predicted according to the prediction model trained by use of the sequence features of the first several *** sequence features include three traditional features and eight newly constructed *** experiments demonstrate that the proposed tracker is superior to most state-of-the-art correlation filter based trackers and has a stable improvement compared to the basic trackers.

关键词： Adaptive padding context information correlation filter(CF) feature group fusion robust visual tracking

来源：评论

学校读者我要写书评

暂无评论

Dehazing Network Based on U-Net Structure and Residual Block 41

Dehazing Network Based on U-Net Structure and Residual Block

引用

第41届中国控制会议

作者： Wenbo Zhang Haoze Li Han Sun Yongqiang Bai School of Automation Beijing Institute of Technology Key Laboratory of Complex System Intelligent Control and Decision Beijing Institute of Technology

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

With the development of deep learning technology,the dehazing method based on convolutional neural network has also developed ***,it still faces some problems such as incomplete dehazing and difficulty in detail *** at the problems,we propose a dehazing network based on U-Net structure and residual *** the down sampling process,we introduce the residual block of attention mechanism,which effectively improves the feature extraction and expression ability of the *** the middle part of the network,the smooth dilation convolution residual block is used to expand the receptive field of the network and improve the restoration effect of small *** the same time,we introduce a feature fusion mechanism based on attention mechanism,which can reduce the loss of information and weight the down-sampling features into intermediate *** the process of up sampling,the method of adding elements instead of feature stitching is used to reduce the number of parameters and realize the function of jump ***,dense convolution residual blocks are added to the jump connection between input and output to further improve the *** experimental results show that the PSNR of this method can reach 32.893,which is better than the existing methods.

关键词： Image Dehazing Residual Network Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Two Degree of Freedom Active Disturbance Rejection control Strategy for Time Delay Systems 42

Two Degree of Freedom Active Disturbance Rejection Control S...

引用

42nd Chinese control Conference, CCC 2023

作者： Shan, Xiyuan Cai, Tao School of Automation Beijing Institute of Technology Beijing100081 China State Key Laboratory of Intelligent Control and Decision of Complex Systems Beijing100081 China

ISBN: (纸本)9789887581543

This paper is concerned with the controller design and the theoretical analysis for time-delay systems, a two degree of freedom (feedforward and feedback) control method is proposed, which combines advantages of the Smith predictor and the active disturbance rejection control (ADRC). The feedforward part of controller is used to track the set point, the feedback part of controller (ADRC) is used to suppress interferences and the Smith predictor is used to correct time delay. The proposed control design is easy to tune parameters and has been proved to effectively control systems with large time delay. The bounded input bounded output (BIBO) stability of closed-loop system is verified. Finally, numerical simulations show the effectiveness and practicality of the proposed design. © 2023 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Time delay

来源：评论

学校读者我要写书评

暂无评论

Detection and Positioning of Workpiece Grinding Area in Dark Scenes with Large Exposure 1

引用

16th International Conference on intelligent Robotics and Applications, ICIRA 2023

作者： Guo, Zhentao Zhao, Guiyu Bian, Jinyue Ma, Hongbin School of Automation Beijing Institute of Technology Beijing100081 China State Key Laboratory of Intelligent Control and Decision of Complex Systems Beijing100081 China

ISBN: (数字)9789819964987

ISBN: (纸本)9789819964970

Workpiece grinding is a crucial process in the smart manufacturing chain. In order to meet the requirements of industrial precision and relieve heavy work, researchers have developed a vision-based grinding robot. However, the problem of workpiece grinding area detection and positioning is difficult to be solved in dark scenes with large exposure. This paper proposes a method that fuses technologies such as processing of the image, coordinate and point cloud, which can accurately detect and locate the workpiece grinding area. Firstly, A method based on YOLOv7 and improved image preprocessing is used to detect labels of the grinding area. Secondly, A model for the prediction of the coordinates based on multiple linear regression was used to predict the coordinates of missing labels for the same grinding area. Finally, Data processing of the point cloud and transformation of the system of coordinates are used to achieve the acquisition of the coordinates of positioning vertices in the grinding area and conversion from the camera coordinate system to the world coordinate system. We used several sets of data for evaluation in our experiments, and the experimental results show that our proposed method can effectively detect the workpiece grinding area. At the same time, our method can also predict the coordinates of missing labels, which provides a more stable and reliable guarantee for industrial production. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Multiple linear regression

来源：评论

学校读者我要写书评

暂无评论

CFSeRec: A Contrastive Framework for Sequential Recommendation 42

CFSeRec: A Contrastive Framework for Sequential Recommendati...

引用

42nd Chinese control Conference, CCC 2023

作者： Wang, Tong Dai, Yaping Shao, Shuai School of Automation Beijing Institute of Technology Beijing100081 China State Key Laboratory of Intelligent Control and Decision of Complex Systems Beijing100081 China

ISBN: (纸本)9789887581543

In sequential recommender systems, the main problems are the long-tailed distribution of data and noise interference. A Contrastive Framework for Sequential Recommendation (CFSeRec) is proposed to solve these two problems respectively. Token shuffling and adversarial attack data augmentation methods are used in the framework to improve the quality and quantity of training data, so that the long-tailed problem is mitigated. Through the application of projection head method, the sequence representation becomes more general and robust, rather than just adapted to the task of contrastive learning. Therefore, the impact of noise on sequence recommender systems is effectively alleviated. Experiments on four public datasets show that CFSeRec achieves state-of-the-art performance in the metrics of hit ratio and normalized discounted cumulative gain, when comparing to the seven previous frameworks. © 2023 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Towards Pose Estimation for Large UAV in Close Range 37

Towards Pose Estimation for Large UAV in Close Range

引用

37th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2022

作者： Ou, Ni Wang, Junzheng Liu, Shangfei Li, Jiehao State Key Laboratory of Intelligent Control and Decision of Complex Systems Beijing Institute of Technology Beijing China

ISBN: (纸本)9781665465366

This paper deals with the problem of 4D pose estimation for large unmanned aerial vehicles (UAVs) in close range. A sensor system consisting of one single point laser range-finder and two cameras is designed and a novel pose estimation method based on vision fusion and point cloud registration is proposed. Our approach works on one-shot mode and only requires 10 samples with real poses for template construction. Through V-rep simulation environment, we generate two 200-sample datasets of different difficulty for evaluation. Error quantiles, 5cm5deg and 10cml0deg are three evaluation metrics used in our ablation experiments. It is illustrated that our method outperforms in robustness and precision due to proposed dimension extension modification and fusion of vision sensors. © 2022 IEEE.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

Deep Multimodal Learning and Fusion Based intelligent Fault Diagnosis Approach

引用

Journal of Beijing Institute of Technology 2021年第2期30卷 172-185页

作者： Huifang Li Jianghang Huang Jingwei Huang Senchun Chai Leilei Zhao Yuanqing Xia Key Laboratory of Complex System Intelligent Control and Decision Beijing Institute of TechnologyBeijing 100081China

Industrial Internet of Things(IoT)connecting society and industrial systems represents a tremendous and promising paradigm *** IoT,multimodal and heterogeneous data from industrial devices can be easily collected,and further analyzed to discover device maintenance and health related potential knowledge *** data-based fault diagnosis for industrial devices is very helpful to the sustainability and applicability of an IoT *** how to efficiently use and fuse this multimodal heterogeneous data to realize intelligent fault diagnosis is still a *** this paper,a novel Deep Multimodal Learning and Fusion(DMLF)based fault diagnosis method is proposed for addressing heterogeneous data from IoT environments where industrial devices ***,a DMLF model is designed by combining a Convolution Neural Network(CNN)and Stacked Denoising Autoencoder(SDAE)together to capture more comprehensive fault knowledge and extract features from different modal ***,these multimodal features are seamlessly integrated at a fusion layer and the resulting fused features are further used to train a classifier for recognizing potential ***,a two-stage training algorithm is proposed by combining supervised pre-training and fine-tuning to simplify the training process for deep structure models.A series of experiments are conducted over multimodal heterogeneous data from a gear device to verify our proposed fault diagnosis *** experimental results show that our method outperforms the benchmarking ones in fault diagnosis accuracy.

关键词： fault diagnosis deep learning multimodal heterogeneous data multimodal fused features

来源：评论

学校读者我要写书评

暂无评论

Tracking and Mapping Strategy for Indoor UAV Based on Entropy Theory:An ORB-SLAM3 Extension 41

Tracking and Mapping Strategy for Indoor UAV Based on Entrop...

引用

第41届中国控制会议

作者： Jian Xu Haoyu Qi Min Xu Yunge Zang Zhen Li Xi Zhang Xiangdong Liu School of Automation and Key Laboratory for Intelligent Control & Decision on Complex Systems Beijing Institute of Technology Beijing Aerospace Automatic Control Institute National Key Laboratory of Science and Technology on Aerospace Intelligence Control

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

For indoor application,which is an entirely GPS-denied environment,visual simultaneous localization and mapping(SLAM) facilitates the real autonomy of unmanned aerial vehicle(UAV) but raises the challenging requirement on accurate localization with low computational *** address this difficulty,a stereo-camera based SLAM system is proposed by applying Entropy theory to *** an extension to ORB-SLAM3,the additional entropy decision module and map processor are specifically *** decision module can improve computing efficiency by deciding whether keyframes or extra optimization should be ***,the map processor is targeted at loading and maintaining the prior map whenever *** results in indoor laboratory environment show that the developed system can achieve the superior localization accuracy in more efficient computation manner with smaller size of mapping compared with ***,the map can be effectively expanded and corrected even when prior information is invalid,greatly increasing the robustness of SLAM system.

关键词： UAV stereo-camera ORB-SLAM3 entropy theory prior map

来源：评论

学校读者我要写书评

暂无评论

Distributed Solver for Discrete-Time Lyapunov Equations over Dynamic Networks with Linear Convergence Rate

引用

IEEE Transactions on Cybernetics 2022年第2期52卷 937-946页

作者： Jiang, Xia Zeng, Xianlin Sun, Jian Chen, Jie Key Laboratory of Intelligent Control and Decision of Complex Systems School of Automation Beijing Institute of Technology Beijing China Beijing Advanced Innovation Center for Intelligent Robots and Systems Beijing Institute of Technology Beijing China

The problem of solving discrete-Time Lyapunov equations (DTLEs) is investigated over multiagent network systems, where each agent has access to its local information and communicates with its neighbors. To obtain a solution to DTLE, a distributed algorithm with uncoordinated constant step sizes is proposed over time-varying topologies. The convergence properties and the range of constant step sizes of the proposed algorithm are analyzed. Moreover, a linear convergence rate is proved and the convergence performances over dynamic networks are verified by numerical simulations. © 2013 IEEE.

关键词： Convex optimization

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：