检索结果-内蒙古大学图书馆

Enhanced Parallel sparse-MLP for Monocular Depth Estimation of Autonomous UAV

Journal of Institute of Control, Robotics and systems 2023年第11期29卷 928-935页

作者： Park, Cheol-Hoon Choi, Hyun-Duck Department of ICT Convergence System Engineering Chonnam National University Korea Republic of

Estimating a high-quality depth map from a single RGB image is a challenging task due to its ill-posed nature. Recently, two dominant trends in computer vision have been the subject of extensive research: attention mechanisms and multi-layer perceptron (MLP)-based vision models. Attention mechanisms, especially multi-head attention (MHA), have demonstrated significant improvements in depth estimation. MHA excels in capturing long-distance information and pixel relationships, yet its complexity quadratically increases with spatial resolution. Consequently, applying MHA to unmanned aerial vehicles with limited hardware resources is infeasible. In contrast, MLP-based vision models offer faster inference due to their linear computational complexity concerning spatial resolution. However, the inherent weakness of the MLP’s inductive bias can hinder generalization without a substantial amount of data. Moreover, the absence of location-dependent local dependencies can hinder the precise estimation of locally detailed depth maps. To address these challenges, this study introduces a novel module called EPsMLP (Enhanced Parallel sparse-MLP), which consists of three parallel branches, including sparse-MLP, local sparse attention, and channel attention. This module can capture global and local dependencies while benefiting from the inductive bias on locality. Furthermore, multi-scale convolutions are used to extract context at various scales for diverse objects. The architecture adopts an encoder-decoder-based structure, incorporating a pre-trained DenseNet-121 encoder. Experimental evaluations were conducted using the NYU-Depth-V2 and KITTI datasets, which are commonly used in monocular depth estimation. The extensive results demonstrate that our network is more efficient and effective than previously proposed methods. © ICROS 2023.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Vessel-position Tracking in 3-sphere Space Using an Interacting Multiple Model Based on Lie Theory

引用

Journal of Institute of Control, Robotics and systems 2024年第6期30卷 643-649页

作者： Jeong, Da Bin Ko, Nak Yong Department of Electronic Engineering Interdisciplinary Program in IT-Bio Convergence System Chosun University Korea Republic of

This paper proposes a method for tracking vessel positions using an interacting multiple model approach based on Lie theory in 3-sphere space. Attitude increments and errors are evaluated and corrected using Lie algebra, as well as exponential and logarithmic operations. The covariance of attitudes is appropriately represented in Lie algebra. The proposed method is compared with two methods, i.e., the unscented Kalman filter (UKF) in which the attitude is represented by a unit quaternion (quaternion-utilizing UKF) and the UKF with a single model that uses Lie theory (Lie theory-incorporating UKF). The quaternion-utilizing UKF represents the attitude using quaternions without incorporating Lie theory, whereas the Lie theory-incorporating UKF represents the attitude as a unit quaternion and utilizes Lie theory. The comparisons indicate that the proposed method effectively reduces tracking errors. Additionally, the proposed method adapts to changes in maneuvering conditions using optimized model probabilities, ensuring timely and appropriate responses. The robustness and effectiveness of the proposed method in various maneuvering scenarios demonstrate its potential for enhancing vessel tracking accuracy in real-world applications. © ICROS 2024.

关键词： Exponential functions

来源：评论

学校读者我要写书评

暂无评论

Graph Neural Network for Digital Twin Network: A Conceptual Framework 6

Graph Neural Network for Digital Twin Network: A Conceptual ...

引用

6th International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2024

作者： Isah, Abubakar Shin, Hyeju Aliyu, Ibrahim Sulaiman, Rashid Muhammad Kim, Jinsul Chonnam National University Department of ICT Convergence System Engineering Gwangju Korea Republic of

ISBN: (纸本)9798350344349

Graph Neural Networks (GNNs) have emerged as a powerful framework for analyzing and extracting information from complex network data. In the realm of Digital Twin Networks (DTN), where physical entities are mirrored in a virtual environment, GNNs offer a transformative approach by leveraging the inherent structure and relationships within digital twins. GNNs enable enhanced data representation and predictive modeling. DTNs encompass several core elements that naturally conform to a graph-like structure, including aspects like network topology and routing patterns. In this paper, we review the concept of graph neural network models, network of digital twin applications, and their comparison with other different fields. © 2024 IEEE.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Designing Data Pipeline for Network Data Management in Digital Twin Network Environment 6

Designing Data Pipeline for Network Data Management in Digit...

引用

6th International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2024

作者： Shin, Hyeju Oh, Seungmin Lee, Jihoon Chung, Gwangmoo Kim, Jinsul Chonnam National University Department of ICT Convergence System Engineering Gwangju Korea Republic of

ISBN: (纸本)9798350344349

With the advancement of Internet of Things (IoT) and 5G technology, new network services such as Virtual Reality/Augmented Reality (VR/AR) are emerging, leading to the expansion of network scale and increasing network traffic. As a solution to the increasing complexity of network operation and maintenance, Digital Twin Network (DTN) technology has been introduced. In order to provide visibility into various aspects for network management, observation and analysis of network data are necessary. Therefore, this paper presents the types of data available in the digital twin network and the lifecycle process of data pipelines. This work will help to effectively manage data internally when implementing the Digital Twin Network in the future. © 2024 IEEE.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

Brands Logo Recognition with Semi-Supervised Learning Using a Robust YOLOv8 Detection Model 6

Brands Logo Recognition with Semi-Supervised Learning Using ...

引用

6th International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2024

作者： Usman, Ali Oh, Seungmin Lee, Junghoon Kim, Jinsul Chonnam National University ICT Convergence System Engineering Department Gwangju Korea Republic of

ISBN: (纸本)9798350344349

Semi-supervised object recognition has emerged as a prominent area of research within computer vision. It offers the potential to significantly decrease the necessity for costly bounding-box annotations. Although there has been considerable success, the current advancements primarily concentrate on two-stage detection networks such as Faster RCNN, while study pertaining to single-stage detectors receives limited attention. This paper centers its attention on semi-supervised learning applied to the advanced and widely adopted single-stage detection network YOLOv8. Our method uses only a minimal amount of labeled data to successfully carry out the training, where we have explored various approaches like data augmentation, student-Teacher network, pseudo labeling, and transfer learning. Furthermore, we have refined YOLOv8 implementation to optimize the advantages offered by semi-supervised learning. To validate our approach, we performed extensive experiments on the challenging OpenLogo Dataset, which contains 27,000 images across a total of 352 classes. The results were obtained with a limited amount of labeled data and a substantial amount of unlabeled data. © 2024 IEEE.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Vessel Location Tracking by Using Unscented Kalman Filter Implemented in Nonlinear Attitude Space

引用

Journal of Institute of Control, Robotics and systems 2023年第4期29卷 347-354页

作者： Jeong, Da Bin Ko, Nak Yong Choi, Hyun-Taek Department of Electronic Engineering Interdisciplinary Program in IT-Bio Convergence System Chosun University Korea Republic of Korea Republic of

In this study, Lie group theory was used for implementing the unscented Kalman filter (UKF). The UKF was used to track an unmanned surface vehicle (USV) by using radar measurements and the data recorded by the automatic identification system. The attitude of the USV was represented using unit quaternions constituting a Lie group called the 3-sphere space (S3). The increment and difference in the attitudes in S3 constitute a Lie algebra. The correspondence between the Lie group component and the Lie algebra component was used by the exponential and logarithm functions of the UKF. Simulation studies verify that the Lie group based UKF results in better convergence property in Kalman gain, measurement innovation, and covariance, as well as better tracking accuracy than the traditional UKF. © ICROS 2023.

关键词： Lie groups

来源：评论

学校读者我要写书评

暂无评论

Beamspace AIC & MDL Algorithm for Based on Concentric Circular Array Antenna 13

Beamspace AIC & MDL Algorithm for Based on Concentric Circul...

引用

Joint 13th International Conference on Soft Computing and Intelligent systems and 25th International Symposium on Advanced Intelligent systems, SCIS and ISIS 2024

作者： Park, Heui-Seon Hwang, Suk-Seung Chosun University Interdisciplinary Program in IT-Bio Convergence System Department of Electronic Engineering Gwangju Korea Republic of School of Electronic Engineering Chosun University Interdisciplinary Program in IT-Bio Convergence System Gwangju Korea Republic of

ISBN: (纸本)9798350373332

In modern wireless communication systems using multiple antenna elements, accurately identifying the signals contained in the received signal is very important. Conventional signal number estimation algorithms operate for the entire set of antenna elements, resulting in high computational complexity. In order to enhance this high computational complexity problem, this paper proposes beamspace based Akaike Information Criterion (AIC) and Minimum Description Length (MDL) algorithms based on a concentric circular array (CCA) antenna for efficient signal number estimation. In addition, computer simulation results are presented to evaluate the signal-number estimation performance of the proposed algorithm. © 2024 IEEE.

关键词： Akaike information criteria Beam space Concentric circular array antenna Minimum description length

来源：评论

学校读者我要写书评

暂无评论

Three-Dimensional Dead-Reckoning Using Lie Theory 24

Three-Dimensional Dead-Reckoning Using Lie Theory

引用

24th International Conference on Control, Automation and systems, ICCAS 2024

作者： Jeong, Da Bin Yong Ko, Nak Chosun University Interdisciplinary Program in IT-Bio Convergence System Graduate School Department of Electronic Engineering Gwangju61452 Korea Republic of Chosun University Interdisciplinary Program in IT-Bio Convergence System Department of Electronic Engineering Gwangju61452 Korea Republic of

ISBN: (纸本)9788993215380

This paper proposes a Dead-reckoning (DR) method for vehicles using Lie theory. This approach treats the pose (position and attitude) and velocity of the vehicle as elements of the Lie group SE2(3) and follows computations based on Lie theory. Previously widely used DR methods suffer from cumulative errors over time due to inaccuracies in the calculated changes from velocity during the motion of the vehicle or small errors in modeling assumptions. By treating the pose and velocity of the vehicle as elements of the Lie group, the proposed method allows for accurate solutions without errors introduced by linearization. Experimental results confirm that the proposed method improves the accuracy of DR, particularly as the motion prediction time interval of the vehicle increases. © 2024 ICROS.

关键词： Exponential functions

来源：评论

学校读者我要写书评

暂无评论

Optimal Cyber Attack Strategy Using Reinforcement Learning Based onCommon Vulnerability Scoring system

引用

Computer Modeling in engineering & Sciences 2024年第11期141卷 1551-1574页

作者： Bum-Sok Kim Hye-Won Suk Yong-Hoon Choi Dae-Sung Moon Min-Suk Kim Department of Electronic Information System Engineering Sangmyung UniversityCheonan31066Republic of Korea Department of Human Intelligence&Robot Engineering Sangmyung UniversityCheonan31066Republic of Korea Intelligent Convergence Research Laboratory Electronics and Telecommunications Research InstituteDaejeon34129Republic of Korea

Currently,cybersecurity threats such as data breaches and phishing have been on the rise due to the many differentattack strategies of cyber attackers,significantly increasing risks to individuals and *** technologies such as intrusion detection have been developed to respond to these cyber ***,advanced integrated cybersecurity that incorporates Artificial Intelligence has been the *** this paper,wepropose a response strategy using a reinforcement-learning-based cyber-attack-defense simulation tool to addresscontinuously evolving cyber ***,we have implemented an effective reinforcement-learning-basedcyber-attack scenario using Cyber Battle Simulation,which is a cyber-attack-defense *** scenarioinvolves important security components such as node value,cost,firewalls,and ***,we applieda new vulnerability assessment method based on the Common Vulnerability Scoring *** approach candesign an optimal attack strategy by considering the importance of attack goals,which helps in developing moreeffective response *** attack strategies are evaluated by comparing their performance using a variety ofReinforcement Learning *** experimental results show that RL models demonstrate improved learningperformance with the proposed attack strategy compared to the original *** particular,the success rateof the Advantage Actor-Critic-based attack strategy improved by 5.04 percentage points,reaching 10.17%,whichrepresents an impressive 98.24%increase over the original ***,the proposed method canenhance security and risk management capabilities in cyber environments,improving the efficiency of securitymanagement and significantly contributing to the development of security systems.

关键词： Reinforcement learning common vulnerability scoring system cyber attack cyber battle simulation

来源：评论

学校读者我要写书评

暂无评论

Optimization Algorithm for Physical Layer Security in Multiple Flying RISs-Based systems

Optimization Algorithm for Physical Layer Security in Multip...

引用

2024 International Conference on electronics, Information, and Communication, ICEIC 2024

作者： Sim, Yuna Sin, Seungseok Ma, Jina Kim, Kyunam Liu, Huaping Moon, Sangmi Hwang, Intae Chonnam National University Deptartment of ICT Convergences System Engineering Gwangju Korea Republic of Alps Electric Korea Co. Ltd. Gwangju Korea Republic of Oregon State University United States Korea Nazarene University Department of IT Artificial Intelligence Cheonan Korea Republic of Chonnam National University Department of Electronic Engineering Department of ICT Convergence System Engineering Gwangju Korea Republic of

ISBN: (纸本)9798350371888

Recently, as data demand has increased due to the rapidly increasing demand for wireless devices and the influence of data traffic, various technologies are being developed to support it. Among them, in the case of millimeter wave (mmWave) frequencies with rich spectra and high data transmission rates, there is a problem of large path loss. As a result, there is a growing interest in unmanned aerial vehicle (UAV) and reconfigurable intelligent surfaces (RIS) that can be reconstructed advantageously to communicate wireless communication environments. Therefore, this paper introduces the penalty-based successive convex approximation (SCA) and minorization-maximization (MM) algorithms to optimize transmission beamforming vectors and RIS beamforming vectors in a large system consisting of multiple users, eavesdroppers, and a number of Flying RISs combining UAV and RIS. © 2024 IEEE.

关键词： Beamforming

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：