检索结果-内蒙古大学图书馆

A Multiscale Gradient Fusion Method for Edge Detection in Color Images Utilizing the CBM3D Filter

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Zhuoyue Tao, Yiyi Ma, Danqing Chen, Jiajing Department of Electrical Engineering and Computer Sciences University of California Berkeley Berkeley United States Department of Computer Science Johns Hopkins University Baltimore United States School of Informatics Computing and Cyber Systems Northern Arizona University Flagstaff United States

In this paper, a color edge detection strategy based on collaborative filtering combined with multiscale gradient fusion is proposed. The block-matching and 3D (BM3D) filter are used to enhance the sparse representation in the transform domain and achieve the effect of denoising, whereas the multiscale gradient fusion makes up for the defect of loss of details in single-scale edge detection and improves the edge detection resolution and quality. First, the RGB images in the dataset are converted to XYZ color space images through mathematical operations. Second, the colored block-matching and 3D (CBM3D) filter are used on the sparse images and to remove noise interference. Then, the vector gradients of the color image and the anisotropic Gaussian directional derivative of the two scale parameters are calculated and averaged pixel-by-pixel to obtain a new edge strength map. Finally, the edge features are enhanced by image normalization and non-maximum suppression technology, and on that basis, the edge contour is obtained by double threshold selection and a new morphological refinement method. Through an experimental analysis of the edge detection dataset, the method proposed has good noise robustness and high edge quality, which is better than the Color Sobel, Color Canny, SE and Color AGDD as shown by the PR curve, AUC, PSNR, MSE, and FOM indicators. Copyright © 2024, The Authors. All rights reserved.

关键词： Collaborative filtering

A leader-follower game theoretic approach to arrest cascading failure in smart grid

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Roy, Sohini Sen, Arunabha School of Computing Informatics and Decision System Engineering Arizona State University AZTempe-85281 United States

—The Smart Grid System (SGS) is a joint network comprising the power and the communication network. In this paper, the underlying intra-and-interdependencies between entities for a given SGS is captured using a dependency model called Modified Implicative Interdependency Model (MIIM) [1]. Given an integer K, the K-contingency list problem gives the list of K-most critical entities, failure of which maximizes the network damage at the current time. The problem being NP complete [2] and owing to the higher running time of the given Integer Linear Programming (ILP) based solution [3], a much faster heuristic solution to generate an event driven self-updating K-contingency list [4] is also given in this paper. Based on the contingency lists obtained from both the solutions, this paper proposes an adaptive entity hardening technique based on a leader-follower game theoretic approach that arrests the cascading failure of entities in the SGS after an initial failure of entities. The validation of the work is done by comparing the contingency lists using both types of solutions, obtained for different K values using the MIIM model on a smart grid of IEEE 14-Bus system with that obtained by simulating the smart grid using a co-simulation system formed by MATPOWER and Java Network Simulator (JNS). The K-contingency list obtained for a smart grid of IEEE 14-Bus system also indicate that the network damage predicted by both the ILP based solution and heuristic solution using MIIM are more realistic compared to that obtained using another dependency model called Implicative Interdependency Model (IIM) [2]. Advantage of using the MIIM based heuristic solution is also shown in this paper when larger SGS of IEEE 118-Bus is considered. Finally, it is shown how the adaptive hardening helps in improving the network performance. Copyright © 2021, The Authors. All rights reserved.

关键词： Smart power grids

A self-updating K-contingency list for smart grid system

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Roy, Sohini Sen, Arunabha School of Computing Informatics and Decision System Engineering Arizona State University AZTempe-85281 United States

A reliable decision making by the operator in a smart grid is contingent upon correct analysis of intra-and-interdependencies between its entities and also on accurate identification of the most critical entities at a given point of time. A measurement based self-updating contingency list can provide real-time information to the operator about current system condition which can help the operator to take the required action. In this paper, the underlying intra-and-interdependencies between entities for a given power-communication network is captured using a dependency model called Modified Implicative Interdependency Model (MIIM) [1]. Given an integer K, the event-driven self-updating contingency list problem gives the list of K-most critical entities, failure of which maximizes the network damage at the current time. Owing to the problem being NP complete, a fast heuristic method to generate a real-time contingency list using system measurements is provided here. The validation of the work is done by comparing the contingency list obtained for different K values using the MIIM model on a smart grid of IEEE 14-Bus system with that obtained by simulating the smart grid using a co-simulation system formed by MATPOWER and Java Network Simulator (JNS). The results also indicate that the network damage predicted by both the ILP based solution [2] and the proposed heuristic solution using MIIM are more realistic compared to that obtained using another dependency model called Implicative Interdependency Model (IIM) [3]. Copyright © 2021, The Authors. All rights reserved.

关键词： Heuristic methods

PPO-Based Energy Efficiency Maximization For RIS-Assisted Multi-User Miso systems 100

学校读者我要写书评

暂无评论

PPO-Based Energy Efficiency Maximization For RIS-Assisted Mu...

100th IEEE Vehicular Technology Conference, VTC 2024-Fall

作者： Iqbal, Amjad Al-Habashna, Ala'a Wainer, Gabriel Boudreau, Gary Bouali, Faouzi Carleton University Department of Systems and Computer Engineering 1125 Colonel By Dr. OttawaONK1S 5B6 Canada Al Hussein Technical University School of Computing and Informatics Amman Jordan Ericsson Canada 349 Terry Fox Dr. KanataONK2K 2V6 Canada Coventry University CoventryCV6 5NU United Kingdom

ISBN: (纸本)9798331517786

In this paper, we explore the integration of a reconfigurable intelligent surface (RIS) with a multi-antenna base station (BS) for downlink multi-user multiple-input-single-output (MU-MISO) systems. We aim to enhance energy efficiency (EE) by jointly optimizing beamforming and phase shifts at the BS and RIS, respectively, while ensuring each mobile user meets their link budget requirements. The resulting optimization problem is inherently non-convex. To address this challenge, we employ proximal policy optimization (PPO), known for efficiently managing non-convex problems and reducing training overhead in continuous action spaces through a clip factor. Furthermore, by leveraging deep neural networks (DNN), the proposed PPO-based solution provides the optimum values for the beamforming at the BS and the phase shift at the RIS, respectively. Finally, we demonstrate the effectiveness and accuracy of the proposed PPO-based algorithm through an extensive simulation campaign, comparing its performance against baseline methods (i.e., fractional programming (FP) and deep deterministic policy gradient (DDPG)). The results show that our proposed PPO-based algorithm outperforms the considered baseline approaches (i.e., FP and DDPG) in terms of EE by 34.2% and 15.8%, respectively. © 2024 IEEE.

关键词： Budget control

Improving Imbalanced Learning by Pre-finetuning with Data Augmentation 4

学校读者我要写书评

暂无评论

Improving Imbalanced Learning by Pre-finetuning with Data Au...

4th International Workshop on Learning with Imbalanced Domains: Theory and Applications, LIDTA 2022

作者： Shi, Yiwen Aslani, Taha Valizadeh Wang, Jing Ren, Ping Zhang, Yi Hu, Meng Zhao, Liang Liang, Hualou College of Computing and Informatics Drexel University Philadelphia United States College of Engineering Drexel University Philadelphia United States Office of Research and Standards Office of Generic Drugs Center for Drug Evaluation and Research Food and Drug Administration Silver Spring United States School of Biomedical Engineering Science & Health Systems Drexel University Philadelphia United States

Imbalanced data is ubiquitous in the real world, where there is an uneven distribution of classes in the datasets. Such class imbalance poses a major challenge for modern deep learning, even with the typical class-balanced approaches such as re-sampling and re-weighting. In this work, we introduced a simple training strategy, namely pre-finetuning, as a new intermediate training stage in between the pretrained model and finetuning. We leveraged the idea of data augmentation to learn an initial representation that better fits the imbalanced distribution of the domain task during the pre-finetuning stage. We tested our method on manually contrived imbalanced datasets (both two-class and multi-class) and the FDA drug labeling dataset for ADME (i.e., absorption, distribution, metabolism, and excretion) classification. We found that, compared with standard single-stage training (i.e., vanilla finetuning), our method consistently attains improved model performance by large margins. Our work demonstrated that pre-finetuning is a simple, yet effective, learning strategy for imbalanced data. © 2022 Y. Shi, T. ValizadehAslani, J. Wang, P. Ren, Y. Zhang, M. Hu, L. Zhao & H. Liang.

关键词： Classification (of information)

Identification and mitigation of false data injection using multi state implicative interdependency model (MSIIM) for smart grid

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Roy, Sohini Sen, Arunabha School of Computing Informatics and Decision System Engineering Arizona State University TempeAZ85281 United States

Smart grid monitoring, automation and control will completely rely on PMU based sensor data soon. Accordingly, a high throughput, low latency Information and Communication Technology (ICT) infrastructure should be opted in this regard. Due to the low cost, low power profile, dynamic nature, improved accuracy and scalability, wireless sensor networks (WSNs) can be a good choice. Yet, the efficiency of a WSN depends a lot on the network design and the routing technique. In this paper a new design of the ICT network for smart grid using WSN is proposed. In order to understand the interactions between different entities, detect their operational levels, design the routing scheme and identify false data injection by particular ICT entities, a new model of interdependency called the Multi State Implicative Interdependency Model (MSIIM) is proposed in this paper, which is an updated version of the Modified Implicative Interdependency Model (MIIM) [1]. MSIIM considers the data dependency and operational accuracy of entities together with structural and functional dependencies between them. A multipath secure routing technique is also proposed in this paper which relies on the MSIIM model for its functioning. Simulation results prove that MSIIM based False Data Injection (FDI) detection and mitigation works better and faster than existing methods. Copyright © 2021, The Authors. All rights reserved.

关键词： Wireless sensor networks

Physics-informed Cyber-Attack Detection in Wind Farms

学校读者我要写书评

暂无评论

Physics-informed Cyber-Attack Detection in Wind Farms

GLOBECOM 2022 - 2022 IEEE Global Communications Conference

作者： Faris Alotibi David Tipper Department of Informatics and Networked Systems School of Computing and Information University of Pittsburgh Pittsburgh PA USA School of Computer Science and Engineering Taibah University Medina Saudi Arabia

ISBN: (纸本)9781665435413

The growth of wind power generation is accelerating worldwide. However, wind turbines need to be carefully controlled to maintain structural stability and optimal power generation. The state of the turbine and power generated is digitized and transmitted over communication networks for physical component monitoring, control, and data management. Motivated by creating power outages, financial loss, and physical wind turbine damage, attackers could craft and inject falsified network traffic to maliciously influence control actions. In this paper, we propose a physics-informed machine learning anomaly detection approach. The mechanism is utilized at each wind turbine, where generated power signals are validated based on laws of physics and incorporated into machine learning algorithm to detect anomalies. We show that the proposed combined physics and machine learning approach improves the accuracy of detecting falsification attacks compared with using machine learning alone.

关键词： Machine learning algorithms Machine learning Telecommunication traffic Wind farms Wind power generation Wind turbines Structural engineering

Light-Weight RetinaNet for Object Detection on Edge Devices

学校读者我要写书评

暂无评论

Light-Weight RetinaNet for Object Detection on Edge Devices

IEEE World Forum on Internet of Things (WF-IoT)

作者： Yixing Li Akshay Dua Fengbo Ren School of Computing Informatics and Decision Systems Engineering Arizona State University

ISBN: (数字)9781728155036

ISBN: (纸本)9781728155043

This paper aims at reducing computation for Retinanet, an mAP-30-tier network, to facilitate its practical deployment on edge devices for providing IoT-based object detection services. We first validate RetinaNet has the best FLOP-mAP trade-off among all mAP-30-tier network. Then, we propose a light-weight RetinaNet structure with effective computation- accuracy trade-off by only reducing FLOPs in computationally intensive layers. Compared with the most common way of trading off computation with accuracy-input image scaling, the proposed solution shows a consistently better FLOPs-mAP trade-off curve. Light-weight RetinaNet achieves a 0.3% mAP improvement at 1.8x FLOPs reduction point over the original RetinaNet, and gains 1.8x more energy-efficiency on an Intel Arria 10 FPGA accelerator in the context of edge computing. The proposed method potentially can help a wide range of the object detection applications to move closer to a preferred corner for a better runtime and accuracy, while enjoys more energy-efficient inference at the edge.

关键词：

Distributed Charging Coordination for Electric Trucks under Limited Facilities and Travel Uncertainties

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Bai, Ting Li, Yuchao Malikopoulos, Andreas A. Johansson, Karl Henrik Mårtensson, Jonas Integrated Transport Research Lab The Division of Decision and Control Systems The Digital Futures KTH Royal Institute of Technology StockholmSE-10044 Sweden School of Computing and Augmented Intelligence Arizona State University TempeAZ-85281 United States School of Civil and Environmental Engineering Cornell University IthacaNY-14853 United States

In this work, we address the problem of charging coordination between electric trucks and charging stations. The problem arises from the tension between the trucks’ nontrivial charging times and the stations’ limited charging facilities. Our goal is to reduce the trucks’ waiting times at the stations while minimizing individual trucks’ operational costs. We propose a distributed coordination framework that relies on computation and communication between the stations and the trucks, and handles uncertainties in travel times and energy consumption. Within the framework, the stations assign a limited number of charging ports to trucks according to the first-come, first-served rule. In addition, each station constructs a waiting time forecast model based on its historical data and provides its estimated waiting times to trucks upon request. When approaching a station, a truck sends its arrival time and estimated arrival-time windows to the nearby station and the distant stations, respectively. The truck then receives the estimated waiting times from these stations in response, and updates its charging plan accordingly while accounting for travel uncertainties. We performed simulation studies for 1, 000 trucks traversing the Swedish road network for 40 days, using realistic traffic data with travel uncertainties. The results show that our method reduces the average waiting time of the trucks by 46.1% compared to offline charging plans computed by the trucks without coordination and update, and by 33.8% compared to the coordination scheme assuming zero waiting times at distant stations. Copyright © 2024, The Authors. All rights reserved.

关键词： Trucks