检索结果-内蒙古大学图书馆

FADE: A Dataset for Detecting Falling Objects around Buildings in Video

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Tu, Zhigang Gao, Zitao Zhang, Zhengbo Zhou, Chunluan Yuan, Junsong Du, Bo The State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University Wuhan430079 China Ant Group co Ltd. Beijing100020 China The Computer Science and Engineering Department The State University of New York at Buffalo BuffaloNY14260 United States The School of Computer Science Wuhan University Wuhan430072 China

Falling objects from buildings can cause severe injuries to pedestrians due to the great impact force they exert. Although surveillance cameras are installed around some buildings, it is challenging for humans to capture such events in surveillance videos due to the small size and fast motion of falling objects, as well as the complex background. Therefore, it is necessary to develop methods to automatically detect falling objects around buildings in surveillance videos. To facilitate the investigation of falling object detection, we propose a large, diverse video dataset called FADE (FAlling Object DEtection around Buildings) for the first time. FADE contains 1,881 videos from 18 scenes, featuring 8 falling object categories, 4 weather conditions, and 4 video resolutions. Additionally, we develop a new object detection method called FADE-Net, which effectively leverages motion information and produces small-sized but high-quality proposals for detecting falling objects around buildings. Importantly, our method is extensively evaluated and analyzed by comparing it with the previous approaches used for generic object detection, video object detection, and moving object detection on the FADE dataset. Experimental results show that the proposed FADE-Net significantly outperforms other methods, providing an effective baseline for future research. The dataset and code are publicly available at https://***/***/. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

A 0.9–1.4-GHz Q-Enhanced CMOS Tunable Bandpass Filter With Constant Fractional Bandwidth and Two Transmission Zeros

学校读者我要写书评

暂无评论

IEEE Microwave and Wireless Technology Letters 2023年第6期33卷 667-670页

作者： Kun Li Xiang Li Shuangyang Wu Pei-Ling Chi Tao Yang School of Electronic Science and Engineering University of Electronic Science and Technology of China Chengdu China Research Center of Advanced RF Chips and Systems Nanhu Laboratory Jiaxing China Department of Electrical and Computer Engineering National Chiao Tung University Hsinchu Taiwan

A tunable differential $Q$ -enhanced bandpass filter covering 0.9–1.4 GHz with constant fractional bandwidth (FBW) is proposed using 40-nm CMOS technology in this letter. The proposed filter is constructed by two resonators which consist of ON-chip integrated inductors and varactors. The inductors of the two resonators are placed close to each other to form the mainline magnetic coupling of the filter. Besides the mainline coupling, the source port and load port are connected directly by capacitors, forming the source-to-load coupling. With the proposed structure, two transmission zeros (TZs) can be obtained, which can greatly increase the stopband rejection. Meanwhile, nMOS cross-coupled pairs are used as the $Q$ -enhanced cells to reduce the insertion loss (IL) and further improve the filter selectivity. A filter prototype based on 40-nm CMOS process is fabricated to validate the proposed structure. The center frequency (CF) of the proposed filter can be tuned from 0.9 to 1.4 GHz while keeping the FBW constant at 8%.

关键词： Resonator filters Band-pass filters Microwave filters Filtering theory Couplings Equivalent circuits Magnetic separation

HEART: Achieving Timely Multi-Model Training for Vehicle-Edge-Cloud-Integrated Hierarchical Federated Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Yang, Xiaohong Liwang, Minghui Wang, Xianbin Cheng, Zhipeng Hosseinalipour, Seyyedali Dai, Huaiyu Jiao, Zhenzhen School of Informatics Xiamen University Fujian China Department of Control Science and Engineering The National Key Laboratory of Autonomous Intelligent Unmanned Systems Japan Frontiers Science Center for Intelligent Autonomous Systems Ministry of Education Tongji University Shanghai China Department of Electrical and Computer Engineering Western University ON Canada School of Future Science and Engineering Soochow University Suzhou China Department of Electrical Engineering University at Buffalo-SUNY BuffaloNY United States Department of Electrical and Computer Engineering NC State University RaleighNC United States iF-Labs Beijing Teleinfo Technology Company Ltd. CAICT Beijing China

The rapid growth of AI-enabled Internet of Vehicles (IoV) calls for efficient machine learning (ML) solutions that can handle high vehicular mobility and decentralized data. This has motivated the emergence of Hierarchical Federated Learning over vehicle-edge-cloud architectures (VEC-HFL). Nevertheless, one aspect which is underexplored in the literature on VEC-HFL is that vehicles often need to execute multiple ML tasks simultaneously, where this multi-model training environment introduces crucial challenges. First, improper aggregation rules can lead to model obsolescence and prolonged training times. Second, vehicular mobility may result in inefficient data utilization by preventing the vehicles from returning their models to the network edge. Third, achieving a balanced resource allocation across diverse tasks becomes of paramount importance as it majorly affects the effectiveness of collaborative training. We take one of the first steps towards addressing these challenges via proposing a framework for multi-model training in dynamic VEC-HFL with the goal of minimizing global training latency while ensuring balanced training across various tasks—a problem that turns out to be NP-hard. To facilitate timely model training, we introduce a hybrid synchronous-asynchronous aggregation rule. Building on this, we present a novel method called Hybrid Evolutionary And gReedy allocaTion (HEART). The framework operates in two stages: first, it achieves balanced task scheduling through a hybrid heuristic approach that combines improved Particle Swarm Optimization (PSO) and Genetic Algorithms (GA);second, it employs a low-complexity greedy algorithm to determine the training priority of assigned tasks on vehicles. Experiments on real-world datasets demonstrate the superiority of HEART over existing methods. © 2025, CC BY-NC-ND.

关键词： Federated learning

Hidden Convexity-Based Distributed Operation of Integrated Electricity-Gas systems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Rong-Peng Song, Yue Liu, Junhong Wang, Xiaozhe Guo, Jinpeng Hou, Yunhe The Department of Electrical and Computer Engineering McGill University MontrealQCH3A 0E9 Canada The State Key Laboratory of Intelligent Autonomous Systems Frontiers Science Center for Intelligent Autonomous Systems Tongji University Shanghai China The Department of Electrical and Electronic Engineering The University of Hong Kong Hong Kong The Department of Energy and Electrical Engineering Ho-Hai University Jiangsu Nanjing China

This letter proposes a hidden convexity-based method to address distributed optimal energy flow (OEF) problems for transmission-level integrated electricity-gas systems. First, we develop a node-wise decoupling method to decompose an OEF problem into multiple OEF subproblems. Then, we propose a hidden convexity-based method to equivalently reformulate nonconvex OEF subproblems as semi-definite programs. This method differs from any approximation and convexification methods that may incur infeasible solutions. Since all OEF subproblems are originally convex or equivalently convexified, we adopt an ADMM to solve the hidden convexity-based distributed OEF problem with convergence analysis. Test results validate the effectiveness of the proposed method, especially in handling a large number of agents. Copyright © 2024, The Authors. All rights reserved.

关键词： Quadratic programming

Single-Phase and Two-Phase Liquid Immersion Cooling of Data Center Power Supply Units for Heat Capture

学校读者我要写书评

暂无评论

Single-Phase and Two-Phase Liquid Immersion Cooling of Data ...

Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic systems (ITHERM)

作者： Haoyun Qiu Pouya Kabirzadeh Vivek S. Garimella Arjit Bali Keerthivasan Gurusami Kang Joon Lee Andrew Stillwell Todd Salamon Nenad Miljkovic Department of Mechanical Science and Engineering University of Illinois Urbana-Champaign Urbana IL Department of Electrical and Computer Engineering University of Illinois Urbana-Champaign Urbana IL Optical Systems and Device Research Laboratory Nokia Bell Labs Murray Hill NJ

ISBN: (数字)9798350364330

ISBN: (纸本)9798350364347

Data center power supply units (PSU) have high power density due to the increasing power demand and their low profiles. Air cooling proves to be increasingly difficult in mitigating the hot spots, and it also consumes extra energy in fluid delivery, air conditioning, and rejection to the ambience. In this work, we redesign and modify the PSU cooling infrastructure and adopt quiescent immersion cooling. This design removes all active air-cooling elements and enables heat capture. The PSU used in the experiments is a 2 kW 80 Plus Platinum commercial unit (Bel PET2000-12-074RA). The cooling performance was characterized by tracking the steady-state case temperatures of the switching semiconductors. Flutec PP1 was used to study the cooling performance variations with respect to PSU load and capture temperature. Conventional fan-cooled PSU with heat sinks served as the benchmark. Experiments show that two-phase immersion cooling grants an overall enhancement of 4.4× over the original air-cooling approach at 40 °C heat capture. The results suggest that heat rejection from the dielectric fluid to the riding heatsink is the main bottleneck, and further reduction of the thermal resistance would allow heat capture at a temperature closer to the liquid boiling point. The PSU thermal management approach studied here can help lower data center power usage effectiveness (PUE) while increasing PSU power delivery density.

关键词： Power system measurements Cooling Density measurement Thermal resistance Thermomechanical processes Immersion cooling Switches Thermal management Resistance heating Data center power

Conflict-Based Search and Improvement Strategies for Solving a New Lexicographic Bi-Objective Multi-Agent Path Finding Problem

学校读者我要写书评

暂无评论

Conflict-Based Search and Improvement Strategies for Solving...

2022 IEEE International Conference on systems, Man, and Cybernetics, SMC 2022

作者： Li, Siyi Li, Xingyang Zhou, Mengchu Zhao, Ziyan Liu, Shixin Northeastern University College of Information Science and Engineering Shenyang110819 China Northeastern University Foshan Graduate School of Innovation Foshan528311 China Northeastern University Faculty of Robot Science and Engineering Shenyang110819 China New Jersey Institute of Technology Department of Electrical and Computer Engineering NewarkNJ07102 United States Macao Inst. of Systems Engineering and Collaborative Laboratory for Intelligent Science and Systems Macau University of Science and Technology 999078 China

ISBN: (数字)9781665452588

ISBN: (纸本)9781665452588

Multi-Agent Path Finding (MAPF) is an important problem with a variety of applications. Its aim is to find collision-free paths for agents having separate start and goal positions. This work proposes a new lexicographic bi-objective MAPF considering different task types, where agents are divided into two kinds to perform critical and acritical tasks. This is common in practical intelligent warehousing scenarios where a critical/acritical-task-performing agent (called c-agent and a-agent, respectively) may represent a full-load/no-load one or the one conducting urgent/non-urgent tasks. The primary objective is to minimize the sum-of-costs of c-agents, while the secondary objective is to minimize the sum-of-costs of a-agents. Two MAPF algorithms are modified to fit and solve the concerned problem for the first time. Moreover, four improvement strategies are embedded to the proposed algorithms and proved to be effective in solving MAPF problems with different task types. © 2022 IEEE.

关键词： Multi agent systems

Road Signs Recognition by Using YOLOv8 Model

学校读者我要写书评

暂无评论

Road Signs Recognition by Using YOLOv8 Model

International Conference on Intelligent systems: Theories and Applications (SITA)

作者： Abdellah Nabou Atheer L. Salih Ibtissam Ben Rabbaa El-Hassan Abdelwahed Mouaad Ait Ahmed Ouhamou Computer Science Dept Computing Systems Engineering Laboratory (LISI) Cadi Ayyad University Marrakesh Morocco Electrical & Communications Engineering Department DHBW University - Campus Friedrichshafen Germany Computer Science Dept Cadi Ayyad University Marrakesh Morocco

Efficient Road sign recognition is key to improving road safety, navigation and to enhance driver assistance for an intelligent transportation system (ITS). However, achieving efficient road sign and traffic recognition is accompanied by many challenges, in our last few years, several deep learning algorithms have been deployed for responding to these limitations. With the exponential increase in the number of vehicles in Morocco, the need for accurate and real-time detection and classification of road signs has become essential. In this paper, we propose an advanced approach for recognizing Moroccan road signs utilizing the YOLOv8 (You Only Look Once) model that very known for its efficiency and effectiveness in object detection tasks. the model is trained and evaluated under a comprehensive dataset that comprising diverse Moroccan road signs. The test of YOLOv8 model give an encouraging result in different performance metrics such as accuracy, recall, precision and mAP which have respectively an average of 0.95, 0.94, 0.96 and 0.97.

关键词：

6th Place Solution to Google Universal Image Embedding

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Gkelios, Socratis Kastellos, Anestis Chatzichristofis, Savvas A. Department of Electrical and Computer Engineering Democritus University of Thrace Greece Intelligent Systems Laboratory Department of Computer Science Neapolis University Pafos Cyprus

This paper presents the 6th place solution to the Google Universal Image Embedding competition on Kaggle. Our approach is based on the CLIP architecture;a powerful pre-trained model used to learn visual representations from natural language supervision. We also utilized the SubCenter ArcFace loss with dynamic margins to improve the distinctive power of class separability and embeddings. Finally, a diverse dataset has been created based on the test’s set categories and the leaderboard’s feedback. By carefully crafting a training scheme to enhance transfer learning, our submission scored 0.685 on the private leaderboard. Copyright © 2022, The Authors. All rights reserved.

关键词： Embeddings

Visual-Semantics Embedding for Deep Hashing-Based Multi-Label Video Retrieval

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Yue, Yuanhao Zou, Qin Cao, Ling Yu, Hongkai Chen, Chi Li, Na School of Cyber Science and Engineering Wuhan University Wuhan430072 China School of Computer Science Wuhan University Wuhan430072 China Department of Electrical Engineering and Computer Science Cleveland State University Cleveland37312 United States State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University Wuhan430079 China TheArchives of Wuhan University Wuhan University Wuhan430072 China

With the explosive growth of short videos in mobile multimedia systems, there is an urgent need for efficient video retrieval. In the past decade, deep hashing technology has occupied a dominant position in content-based visual retrieval and has achieved remarkable success in image retrieval. However, for video retrieval, existing hashing-based methods still have a lot of room for improvement, especially concerning multi-label videos. There are mainly two reasons: 1) The similarity is not well-defined for multi-label videos. Traditionally, pairwise similarity is defined purely based on the explicit text information of the labels, which ignores the implicit semantic relations between them, leading to inaccurate distance measurements;2) Time-series information of video frames is not properly used in feature extraction. Most methods assign equal weight to each frame in their models, neglecting the fact that the content of a video is often determined by several keyframes. To solve the above problems, a novel video hashing method is proposed in this paper. First, a visual-semantics embedding soft similarity is developed to calculate the distance between pairwise videos, where the implicit semantic association of labels is learned by a graph convolution network (GCN). Second, a hybrid attention module, consisting of a self-attention block and a relation-attention block, is integrated into the hashing network. The attention module assigns different weights to the frames according to their importance in determining the content of the video. Experimental results show that the proposed method achieves significant improvements over competing ones. © 2023, The Authors. All rights reserved.

关键词： Convolution