检索结果-内蒙古大学图书馆

Journal of Network Intelligence 2024年第2期9卷 1119-1133页

作者： Huang, Jia-Yan Xu, Xin-Xin Luo, Shao-Ye Zhao, Wan-Qi Fan, Hao-Yi Grau, Antoni Guerra, Edmundo Wang, Chuan-Sheng Zhang, Fu-Quan New Engineering Industry College Putian University Putian351100 China The University of Sheffield SheffieldS10 2TN United Kingdom School of Computer and Artificial Intelligence Zhengzhou University Zhengzhou450001 China Department of Automatic Control Technical Polytechnic University of Catalonia Barcelona08034 Spain Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Minjiang University Fuzhou350108 China

Image dehazing is an important research topic and hotspot in the fields of image processing and computer vision. Therefore, evaluating the performance of image dehazing algorithms has become an import research issue. However, current image quality assessment (IQA) algorithms only judge the final dehazed image and do not give an analysis of the capabilities of the image dehazing algorithm. In order to fill this gap, we propose an image dehazing quality assessment method that integrated the dark channel prior with image depth estimation. The proposed method first uses the dark channel prior to estimate the absolute amount of haze removed from each pixel as an indication of the performance of dehazing. Then, a dual-scale variance ratio method is introduced to estimate the scene depth of the image. Finally, the relative dehazing is calculated from the obtained absolute dehazing and the image depth to represent the capability of the dehazing algorithm. The method proposed in this paper first introduces the concepts of relative and absolute dehazing quantities, along with pioneering a pixel-level evaluation of dehazing outcomes. Extensive experiments show that the algorithm proposed in this paper is more in line with human subjective judgment than other IQA algorithms. © 2024, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

ResCue: Inferring Fine-Grained Traffic Matrices via Distributed Deep Residual Networks

ResCue: Inferring Fine-Grained Traffic Matrices via Distribu...

引用

International Conference on Network and Service Management

作者： Lorenzo Pappone Cristian Zilli Alessio Sacco Flavio Esposito Department of Computer Science Saint Louis University USA Department of Control and Computer Engineering Politecnico di Torino Italy

ISBN: (数字)9783903176669

ISBN: (纸本)9798331505158

Network measurement and telemetry techniques are central to the management of modern computer networks. Internet traffic matrix estimation is a popular technique employed for network management and telemetry to reconstruct missing information. Existing approaches use statistical methods, which often make impractical assumptions about the structure of the Internet traffic matrix. Data-driven methods, instead, heavily rely on the assumption of full knowledge of network topology data, that may be unavailable or impractical to collect. In this work, we propose ResCue, a deep residual networks technique to infer fine-grained Internet network traffic starting from spatial coarse-grained measurements. To address scenarios with network visibility constraints, we design a federated learning approach for fine-grained traffic prediction with partial network knowledge. Our evaluation across real-world traffic data shows that our proposed approach outperforms existing interpolation techniques and that our federated learning design achieves similar accuracy with respect to its centralized counterpart while requiring only partial knowledge of the network.

关键词： Knowledge engineering Training Interpolation Federated learning Superresolution Telecommunication traffic computer architecture Internet Telemetry Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Two-Stage Robust Optimization Under Decision Dependent Uncertainty

引用

IEEE/CAA Journal of Automatica Sinica 2022年第7期9卷 1295-1306页

作者： Yunfan Zhang Feng Liu Yifan Su Yue Chen Zhaojian Wang João P.S.Catalão State Key Laboratory of Power System and Generation Equipment the Department of Electrical EngineeringTsinghua UniversityBeijing 100084China Department of Mechanical and Automation Engineering the Chinese University of Hong KongHong Kong SARChina Ministry of Education Key Laboratory of System Control and Information Processing the Department of AutomationShanghai Jiao Tong Universityand also with Shanghai Engineering Research Center of Intelligent Control and ManagementShanghai 200240China Faculty of Engineering of the University of Porto and Institute for Systems and Computer Engineering Technology and Science(INESC TEC)Porto 4200-465Portugal IEEE

In the conventional robust optimization(RO)context,the uncertainty is regarded as residing in a predetermined and fixed uncertainty *** many applications,however,uncertainties are affected by decisions,making the current RO framework *** paper investigates a class of two-stage RO problems that involve decision-dependent *** introduce a class of polyhedral uncertainty sets whose right-hand-side vector has a dependency on the here-and-now decisions and seek to derive the exact optimal wait-and-see decisions for the second-stage problem.A novel iterative algorithm based on the Benders dual decomposition is proposed where advanced optimality cuts and feasibility cuts are designed to incorporate the uncertainty-decision *** computational tractability,robust feasibility and optimality,and convergence performance of the proposed algorithm are guaranteed with theoretical *** motivating application examples that feature the decision-dependent uncertainties are ***,the proposed solution methodology is verified by conducting case studies on the pre-disaster highway investment problem.

关键词： Benders decomposition decision-dependent uncertainty endogenous uncertainty robust optimization(RO)

来源：评论

学校读者我要写书评

暂无评论

A Fusion Model for Artwork Identification Based on Convolutional Neural Networks and Transformers

arXiv

引用

arXiv 2025年

作者： Wang, Zhenyu Song, Heng The School of Control and Computer Engineering North China Electric Power University Beijing102206 China

The identification of artwork is crucial in areas like cultural heritage protection, art market analysis, and historical research. With the advancement of deep learning, Convolutional Neural Networks (CNNs) and Transformer models have become key tools for image classification. While CNNs excel in local feature extraction, they struggle with global context, and Transformers are strong in capturing global dependencies but weak in fine-grained local details. To address these challenges, this paper proposes a fusion model combining CNNs and Transformers for artwork identification. The model first extracts local features using CNNs, then captures global context with a Transformer, followed by a feature fusion mechanism to enhance classification accuracy. Experiments on Chinese and oil painting datasets show the fusion model outperforms individual CNN and Transformer models, improving classification accuracy by 9.7% and 7.1%, respectively, and increasing F1 scores by 0.06 and 0.05. The results demonstrate the model’s effectiveness and potential for future improvements, such as multimodal integration and architecture optimization. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

IMC Based PID controller for Spherical Tank Liquid Level control

IMC Based PID Controller for Spherical Tank Liquid Level Con...

引用

AIML-Applications for engineering & Technology (ICAET), International Conference on

作者： C. B. Kadu V. P. Bhosale P. S. Vikhe V. V. Mandhare Department of Instrumentation & Control Engineering Pravara Rural Engineering College Loni India Department of Computer Engineering Pravara Rural Engineering College Loni India

ISBN: (数字)9798350355611

ISBN: (纸本)9798350355628

In the process industries, it's hard to control a non-linear process. Nonlinear behavior is frequently seen in real processes. The challenging problem of controlling a spherical tank is result of its nonlinearity caused by the continuously varying cross section. An IMC-based PID controller is employed to improve efficiency of controlling the level of spherical tanks which exhibit nonlinear dynamics. For processes up to second-order with dead time and first-order numerator dynamics, a single tuning parameter will be offered. The process model is determined employing mathematical modeling and standard step response-based system identification. The laboratory model of an interacting and non-interacting nonlinear system is introduced, which is capable of performing more complex tasks. To verify the effectiveness of the suggested IMC based PID controller, results of real-time experiments and a simulation study on spherical tank level control are incorporated.

关键词： Process control Real-time systems Mathematical models Regulation Nonlinear dynamical systems System identification Reliability Level control Tuning Standards

来源：评论

学校读者我要写书评

暂无评论

An Unmanned Aerial Vehicle-aided sparse data gathering in Wireless Sensor Networks 8

An Unmanned Aerial Vehicle-aided sparse data gathering in Wi...

引用

8th IEEE International Conference on Cloud Computing and Intelligence Systems, CCIS 2022

作者： Lv, Cuicui Wang, Peijin Wang, Peng Du, Zhenbin Li, Xiangming YanTai University School of Computer and Control Engineering YanTai264005 China YanTai University School of Environment and Materials Engineering YanTai264005 China

ISBN: (纸本)9781665477352

Sensory data in Wireless Sensor Networks have many redundancies. For this problem, the paper presents a sparse data gathering framework, which adopts Compressive Sensing(CS) to reduce data redundancies, and utilizes Unmanned Aerial Vehicle (UAV) as the mobile relay to collect the compressed data. In this framework, the UAV is only required to collect data of the fewer nodes to generate the measurement matrix, and joins the covariance matrix to design the sensing matrix in CS. To reduce energy consumption, we use greedy algorithm to plan the route of the UAV. The experiments demonstrate that the proposed framework can make data be sparser and the recovery performance be better. © 2022 IEEE.

关键词： Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Power User Profiling Techniques for Fusing Heterogeneous Data from Multiple Sources 5

Power User Profiling Techniques for Fusing Heterogeneous Dat...

引用

5th International Conference on Smart Grid and Energy engineering, SGEE 2024

作者： Liu, Zehua Wang, Yanqin Liu, Yifan Miao, Hongying School of Control and Computer Engineering North China Electric Power University Beijing China State Grid Jibei Electric Power Co. Ltd. Chengde Power Supply Company Chengde China

ISBN: (纸本)9798331530686

In order to realize the accurate classification and feature grasping of power users by the power supply department, this paper proposes a method of power user portrait by fusing heterogeneous data from multiple sources. The method first uses the K-means++ algorithm and combines the elbow rule to determine the optimal number of clusters K, and clusters the user load data and basic information separately;subsequently, the key load features are screened out by the mRMR method, and the radar diagram is used to visually display the features of different categories of users;finally, the refined construction of the user portrait is realized by combining various types of attribute labels, which provides the basis for the enterprise to provide personalized services. In this paper, the effectiveness of the proposed method is verified on two datasets. © 2024 IEEE.

关键词： cluster analysis feature selection multi-source heterogeneous data power user profiling

来源：评论

学校读者我要写书评

暂无评论

Multi-dimensional networks as a tool to model, analyze, and interpret multi-subject brain connectivity in hyperscanning settings*

Multi-dimensional networks as a tool to model, analyze, and ...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Maria Grazia Puxeddu Greta Rinaldini Laura Astolfi Department of Computer Control and Management Engineering University of Rome La Sapienza Rome Italy

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Social interactions are woven in the fabric of our lives, from birth to adulthood. Studies show that when social connections are lacking, there is a high probability of developing poor mental and physical health. Conversely, some mental disorders result in a compromised ability to form these interactions, with autism being a prime example. For these reasons, understanding the neural mechanisms underlying social interactions is crucial. Hyperscanning experiments, where brain activity is simultaneously recorded from multiple subjects during social tasks, have emerged as a powerful tool for tackling this issue, and the study of brain-to-brain connectivity has provided significant insights into social neuroscience. However, no agreed-upon methods exist for analyzing and interpreting such data, and most of the analyses still rely on techniques designed for single-subject experiments. Here, we propose a framework to model brain-to-brain connectivity through multi-dimensional networks. Under this conceptualization, both intra- and inter-brain connectivity are viewed as components of a higher-order network that can be studied by extending graph theory indices to a multidimensional case. In this article, we also discuss how these indices can be interpreted in the context of multi-subject networks. Future investigations will apply this framework to hyperscanning datasets to provide new insights into the neural circuits associated with social interactions.

关键词： Neuroscience Mental disorders Biological system modeling Neural circuits Brain modeling Weaving Graph theory Fabrics Integrated circuit modeling Bioinformatics

来源：评论

学校读者我要写书评

暂无评论

Retinal Vessel Segmentation Based on UNet and Edge Detection 12

Retinal Vessel Segmentation Based on UNet and Edge Detection

引用

12th International Conference on Information Technology in Medicine and Education, ITME 2022

作者： Yan, Peixin Li, Zuoyong Hu, Rong Cao, Xinrong Fujian University of Technology College of Computer Science and Mathematics Fuzhou350118 China Minjiang University Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Colloge of Computer and Control Engineering Fuzhou350121 China

ISBN: (纸本)9798350310153

Retinal vessel segmentation is important for the diagnosis of many diseases. The segmentation of microscopic vessels is one of the challenges because of the complex structure of retinal vessels, which makes it challenging to tell the vessels apart from the background. In order to improve the segmentation performance of retinal vessels, we use the edge detection method of HED to optimize the segmentation effect of UNet. Using Hybrid Dilated Convolution not only increases the perceptual field, but also avoids the loss of information continuity caused by the continuous use of dilated convolution. The Pyramid Pooling Module is used at the end of the encoder to compensate for the information loss caused by downsampling. Finally, the output of edge detection and image segmentation is combined as the final result. The model is evaluated in two mainstream datasets DRIVE and CHASE-DB1, and the experimental results show that the proposed model can improve the performance of retinal vessel segmentation. © 2022 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Enhancing Learning and Therapy for Children with Autism: Innovative Activities with the NAO Robot

Enhancing Learning and Therapy for Children with Autism: Inn...

引用

E-Health and Bioengineering Conference (EHB)

作者： Iustina-Bianca Bulai Otilia Zvorişteanu Dept. of Computer Science and Engineering Faculty of Automatic Control and Computer Engineering “Gheorghe Asachi” Technical University of Iaşi Iaşi Romania

ISBN: (数字)9798331532147

ISBN: (纸本)9798331532154

Worldwide, autism affects 1 in 100 children, with diagnoses steadily rising since the 2000s. Romania lacks official statistics on autism prevalence, but global estimates suggest similarity. Children with autism struggle with social interactions and verbal communication, prompting interest in social robots for therapy and education. Tailored robot-assisted sessions offer potential for cognitive development, vocabulary enhancement, social interaction, and behavioral stimulation like concentration and attention. The NAO robot's user-friendly design, technical capabilities, and integrated features make it well-suited for these purposes. This paper presents activities that exemplify the concept of “learning through play,” allowing children to engage in exploration, experimentation, and effective learning in an enjoyable manner.

关键词： Autism Vocabulary Hospitals Social robots Education Biomedical engineering

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：