检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第12期5卷 6500-6508页

作者： Chen, Luefeng Cao, Wei Zheng, Biao Wu, Min Pedrycz, Witold Hirota, Kaoru China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Ministry of Education Engineering Research Center of Intelligent Technology for Geo-Exploration Wuhan430074 China University of Alberta Department of Electrical and Computer Engineering EdmontonABT6G 2R3 Canada Macau University of Science and Technology Institute of Systems Engineering Taipa999078 China Istinye University Research Center of Performance and Productivity Analysis Istanbul34010 Turkey Tokyo Institute of Technology Yokohama226-8502 Japan

In the human pose estimation task, on the one hand, 3-D pose always has difficulty in dividing different 2-D poses if the view is limited;on the other hand, it is hard to reduce the lifting ambiguity because of the lack of depth information, it is an important and challenging problem. Therefore, two-stage representation refinement based on the convex combination for 3-D human pose estimation is proposed, in which the two-stage method includes a dense-spatial-temporal convolutional network and a local-to-refine network. The former is applied to determine the features between each video frame;the latter is used to get the different scales of pose details. It aims to address the difficulty of estimating 3-D human pose from 2-D image sequences. In such a way, it can better use the relations between every frame in the sequence of the pose video to produce more accurate results. Finally, we combine the above network with a block called convex combination to help refine the 3-D pose location. We test the proposed approach on both Human3.6m and MPII datasets. The result confirms that our method can achieve better performance than improved CNN supervision, a simple yet effective baseline, and coarse-to-fine volumetric prediction. Besides, a robustness test experiment is carried out for the proposed method while the input is interrupted. The result verifies that our method shows better robustness. © 2024 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

DeepTQ: Predictive TSN Switch Queue Length Based on Deep Learning 1

引用

7th Chinese Conference on Swarm Intelligence and Cooperative control, CCSICC 2023

作者： Jia, Yanxin Xu, Long Xiong, Wei Wang, Xin Shang, Zhijun Yuan, Zijun 3onedata Co. Ltd. Shenzhen China Ministry of Education School of Automation and the Key Laboratory of Measurement and Control of Complex Systems of Engineering Southeast University Nanjing China 3onedata Qitong Co. Ltd. Shanghai China University of Chinese Academy of Sciences Beijing China

ISBN: (数字)9789819733361

ISBN: (纸本)9789819733354

With the rapid development of network technology, TSN (Time-Sensitive Networking) is in the stage of rapid development. It needs to ensure the deterministic transmission of time-sensitive flow, further improve the network throughput, and meet the increasing traffic demand. SDN (Software-Defined Networking) can realize the automatic management of network resources by decoupling the control plane and the data plane. Through global network information, it uses efficient route scheduling to achieve load balancing and improve network throughput. However, the prerequisite for route scheduling is that accurately predict the queue length of TSN switch, so that route scheduling can be carried out according to the predicted results. To address the queue length prediction problem of TSN switch, DeepTQ (Predictive TSN Switch Queue Length Based on Deep Learning) is proposed, which predicts the queue length of TSN switches in the next time slot using deep learning methods. It consists of the feature selection phase and the model prediction phase. In DeepTQ, the feature selection phase uses the Spearman rank correlation coefficient to remove redundant features in the datasets. The model prediction phase uses an improved Transformer Encoder and GRU (Gate Recurrent Unit) to construct the prediction model. Simulation results demonstrate that DeepTQ achieves superior performance compared to existing methods and individual components contribute to its overall effectiveness. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Software defined networking

来源：评论

学校读者我要写书评

暂无评论

ResFFont: Few-Shot Font Generation based on Reversible Network 43

ResFFont: Few-Shot Font Generation based on Reversible Netwo...

引用

43rd Chinese control Conference, CCC 2024

作者： Ju, Chuanying Zhang, Ziying Chen, Xin China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China China University of Geosciences School of Arts and Communication Wuhan430073 China

ISBN: (纸本)9789887581581

The technique of generating fonts with few-shot learning has become a popular research topic. The goal of this technique is to transform the source-to-target domain by using a small number of sample images to generate characters for new fonts. However, current research is limited by model frameworks and migration algorithms, which pose challenges for maintaining content integrity and stroke-level style consistency when generating Chinese characters. This paper proposes ResFFont, a reversible network-based model that uses a unique reversible framework and Global and Local Feature Fusion Module (GaLSFM) to ensure complete delivery of content information and stroke-level style consistency. The experimental results confirm the effectiveness and superiority of the ResFFont model, providing a new solution to the font generation problem. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Few-Shot Font Generation Reversible Network Style Transfer

来源：评论

学校读者我要写书评

暂无评论

An Improved Task Planning Method for Unmanned Swarms with Coupled Physical and Logical Constraints

An Improved Task Planning Method for Unmanned Swarms with Co...

引用

2024 IEEE International Conference on Unmanned systems, ICUS 2024

作者： Chen, Xiubin Zhang, Lei Xu, Fang Yao, Weiran School of Astronautics Harbin Institude of Technology Harbin China Wuhan 2nd Ship Design and Research Institute Wuhan China Marine Military Intelligence Innovation Institute CSSC Systems Engineering Research Institute Beijing China National Key Laboratory of Complex System Control and Intelligent Agent Cooperation Beijing China

ISBN: (纸本)9798350384185

Traditional task allocation methods for unmanned swarm systems ignore the effects of actual paths, resulting in estimation accuracy *** paper formulates task planning problem by incorporating physical and logical constraints, and establishes an integrated framework of task allocation and path ***-based search method is used to address path planning with physical constraints.A genetic algorithm is employed to solve multi-traveling salesman allocation problem.A bounded suboptimal optimization, a data dictionary, and an island model are introduced to accelerate the convergence speed of the genetic *** experiments verify that compared to the decoupled task planning methods, the proposed method improves task execution efficiency and remains adaptability to various complex spatial maps. © 2024 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Hardware-in-the-loop Simulation and Experimental System for Process control of Marine Resources Exploration 43

Hardware-in-the-loop Simulation and Experimental System for ...

引用

43rd Chinese control Conference, CCC 2024

作者： Wang, Zeyi Wang, Chenxuan Ma, Zhejiaqi Zeng, Kanghui Lu, Chengda Wang, Yawu Wu, Min China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China China University of Geosciences School of Future Technology Wuhan430074 China

ISBN: (纸本)9789887581581

Effective control techniques are essential for offshore platform systems to suppress disturbances caused by the complex marine environment and ensure the stable operation of the systems during the exploration of marine resources. This paper establishes a hardware-in-the-loop based marine resource exploration process control simulation and experiment system to simulate the effects of marine environmental factors on the offshore platform. The marine environment is simulated using JONSWAP spectra and real marine environment data. The dynamics of offshore platform is modeled using Lagrange's equation method. And the control effect is demonstrated using virtual reality technology. Ultimately, the dynamic positioning system control experiment, which uses the equivalent input disturbance method, validates the experimental system. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Hardware-in-the-loop simulation

来源：评论

学校读者我要写书评

暂无评论

Effective Indoor Fire Detection with Channel Shuffle Module 21

Effective Indoor Fire Detection with Channel Shuffle Module

引用

5th International Conference on Computer Science and Application engineering, CSAE 2021

作者： Ge, Haotian Cao, Yichao Lu, Xiaobo School of Automation Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China

ISBN: (纸本)9781450389853

In recent years, methods based on computer vision and deep learning become the mainstream approaches in fire detection. However, the expensive computation cost of 3D convolutional neutral network (CNN) is unbearable and it is difficult for them to capture the fire regions of videos in time. In this paper, we design a module named channel shuffle module (CSM) based on 2D CNN to keep the balance between computation cost and accuracy. By fusing RGB frame and differential frame, CSM improves the ability of 2D CNN in temporal information extraction which much less cost than methods based on 3D CNN. Four different structures of CSM are proposed and we choose the best one by experiment results. Also, experiments prove that the performances of TSN and TSM are improved with CSM in sequence classification. The accuracy of TSM with CSM is 99.2045%, false positive rate reaches 0.7890% and false negative rate reaches 0.4530%, which demonstrates the efficiency of CSM in temporal feature modeling.. © 2021 Association for Computing Machinery. All rights reserved.

关键词： Fires

来源：评论

学校读者我要写书评

暂无评论

CDP-GAN:Near-Infrared and Visible Image Fusion Via Color Distribution Preserved GAN

引用

IEEE/CAA Journal of Automatica Sinica 2022年第9期9卷 1698-1701页

作者： Jun Chen Kangle Wu Yang Yu Linbo Luo the School of Automation Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex SystemsEngineering Research Center of Intelligent Technology for Geo-ExplorationMinistry of EducationChina University of GeosciencesWuhan 430074China the Shanghai Institute of Technical Physics Key Laboratory of Infrared System Detecting and Imaging TechnologyChinese Academy of SciencesShanghai 200083China the School of Mechanical Engineering and Electronic Information China University of GeosciencesWuhan 430074China

Dear Editor,This letter is concerned with dealing with the great discrepancy between near-infrared(NIR)and visible(VS)image fusion via color distribution preserved generative adversarial network(CDP-GAN).Different from the global discriminator in prior GAN,conflict of preserving NIR details and VS color is resolved by introducing an attention guidance mechanism into the ***。

关键词： details dealing preserving

来源：评论

学校读者我要写书评

暂无评论

A Two-step Model for Multi-object Tracking 21

A Two-step Model for Multi-object Tracking

引用

5th International Conference on Computer Science and Application engineering, CSAE 2021

作者： Zhang, Shuaishuai Lu, Xiaobo Du, Songlin Southeast University Key Laboratory of Measurement and Control of Complex Systems of EngineeringMinistry of Education School of Automation Nanjing China

ISBN: (纸本)9781450389853

Multi-object tracking is widely used in video analysis. However, due to the limitation of detector performance, many multi-object tracking models have the problem of detecting two objects into one object in some occlusion scenes. In this paper, we propose a twostep model for handling this problem. In the first step model, the non-occlusion targets are detected and embeddings are extracted, while the occlusion areas are identified. The second step model processes the occlusion areas to obtain occlusion targets' accurate positions and embeddings. Finally, we integrate and optimize the output results of the two steps models. Experiments show that the number of false positives and missed positives in our model's object detection is significantly reduced. The multi-object tracking performance (MOTA metric) is improved by nearly 3% compared with other models.. © 2021 Association for Computing Machinery. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Multi Scale Attention Network for Crowd Counting 21

Multi Scale Attention Network for Crowd Counting

引用

5th International Conference on Computer Science and Application engineering, CSAE 2021

作者： Yang, Xiangpeng Lu, Xiaobo Key Laboratory of Measurement and Control of Complex Systems of Engineering School of Automation Southeast University Ministry of Education Nanjing China

ISBN: (纸本)9781450389853

Reasonable management and control of extra crowded scenes have become a hot topic in recent years. Counting people from density map generated from the object location annotations is an effective way to analyze crowd information and control crowds in severely congested scenes. In this paper, we propose a novel endto- end crowd counting method called MSANet for crowd counting. MSANet consists of the VGG16 backbone as the fronted part, two branches as the back-end part, including the attention map extractor to predict crowd states (means with people or not), and density map branch to regress the density map. What is more, to obtain high-resolution density map, we combine different scale maps from the front part to the back-end part. On the design of the loss function, to enhance the resolution of the predicted map and its structural similarity to ground truth, we proposed a new loss function for crowd counting. The test result based on the public dataset ShanghaiTech and Subway Crowd Counting Dataset supported by the Nanjing Metro demonstrates the effectiveness of our method.. © 2021 Association for Computing Machinery. All rights reserved.

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

Skeleton-based Action Recognition Using Two-stream Graph Convolutional Network with Pose Refinement 41

Skeleton-based Action Recognition Using Two-stream Graph Con...

引用

第41届中国控制会议

作者： Biao Zheng Luefeng Chen Min Wu Witold Pedrycz Kaoru Hirota School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Department of Electrical and Computer Engineering University of Alberta Tokyo Institute of Technology

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

With the development of science and technology,graph convolutional network has made great progress in improving the accuracy of action ***,there still exists some deficiencies in current ***,the human skeleton point coordinates entering into the network are barely refined,which may cause large ***,the second-order information(the length and direction of bones),which can reflect action characteristics discriminatively,is rarely *** solve the above issues,a two stream graph convolutional network with pose refinement for skeleton based action recognition is ***,we use an adaptive block to to help improve the *** test our method on Kinetics dataset and the experiment show it can get better results than some recent methods,which plays a positive role in future research.

关键词： Graph convolutional network Pose refinement Skeleton based action recognition Adaptive block

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：