4.2 Article

Joint frame rate adaptation and object recognition model selection for stabilized unmanned aerial vehicle surveillance

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Telecommunications

Starling Flocks-Inspired Resource Allocation for ISAC-Aided Green Ad Hoc Networks

Jiaxing Wang et al.

Summary: This article proposes an environment awareness integrated sensing and communication (ISAC) network association architecture to overcome the low-latency and energy-efficient demand of future green ad hoc networks. The resource allocation optimization problem of the ISAC architecture is efficiently solved using the graph theory and a novel low-complexity algorithm. Simulation results demonstrate that the proposed ISAC network association architecture can achieve lower transmission delay compared to traditional methods, and near-optimal latency performance can be obtained with the optimization method.

IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING (2023)

Article Engineering, Electrical & Electronic

Reliable Transmission for NOMA Systems With Randomly Deployed Receivers

Yibo Zhang et al.

Summary: In this paper, the reliable transmission scheme of downlink Non-Orthogonal Multiple Access (NOMA) systems is investigated. The base station's coverage area is divided into multiple annular regions, with receivers randomly distributed within them, to achieve NOMA pairing. Bit Error Rate (BER) expressions with Quadrature Phase-Shift Keying (QPSK) modulation are derived. The BER performance of the receiver with the worst channel gain in each region is studied to ensure reliable communications. An optimal power allocation algorithm is proposed to minimize the transmission power while meeting a given BER constraint. Extensive simulations validate the accuracy of the obtained BER expressions and the effectiveness of the proposed algorithm. These findings offer valuable insights for achieving reliable transmission in NOMA systems with randomly deployed receivers.

IEEE TRANSACTIONS ON COMMUNICATIONS (2023)

Article Robotics

Visual localization with a monocular camera for unmanned aerial vehicle based on landmark detection and tracking using YOLOv5 and DeepSORT

Liqun Ma et al.

Summary: This study presents a landmark-based visual localization framework for unmanned aerial vehicles flying at low altitudes. YOLOv5 and DeepSORT are utilized for multi-object detection and tracking, and the localization is achieved by comparing geotagged transmission towers with annotated images captured by a monocular camera. The proposed strategy avoids the need for satellite images as a reference map, saving GPU memory and enabling end-to-end implementation on small unmanned aerial vehicles.

INTERNATIONAL JOURNAL OF ADVANCED ROBOTIC SYSTEMS (2023)

Proceedings Paper Computer Science, Artificial Intelligence

ROMA: Run-Time Object Detection To Maximize Real-Time Accuracy

JunKyu Lee et al.

Summary: This paper analyzes the effects of dynamically varying video contents and detection latency on the real-time detection accuracy, and proposes a new run-time accuracy variation model, ROMA. ROMA is designed to select the optimal detector in real time without label information to maximize real-time object detection accuracy. Utilizing four YOLOv4 detectors on an NVIDIA Jetson Nano, ROMA achieves real-time accuracy improvements of 4 to 37% compared to individual YOLOv4 detectors and two state-of-the-art runtime techniques in a scenario of dynamically varying video contents and detection latency consisting of MOT17Det and MOT20Det datasets.

2023 IEEE/CVF WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV) (2023)

Article Computer Science, Information Systems

Workload-Aware Scheduling Using Markov Decision Process for Infrastructure-Assisted Learning-Based Multi-UAV Surveillance Networks

Soohyun Park et al.

Summary: This paper proposes a novel UAV scheduling algorithm that solves the power limitation problem of UAVs through wireless power transfer, and introduces a data delivery scheme that allows UAVs to fairly deliver training data to towers to prevent problems caused by data imbalance. The performance evaluation shows that our proposed algorithm ensures sufficient times for resource exchanges between towers and UAVs, achieves the most even and uniform data collection during the processes compared to other algorithms, and converges the performance of all towers to optimal levels.

IEEE ACCESS (2023)

Article Engineering, Electrical & Electronic

Age of Information in Energy Harvesting Aided Massive Multiple Access Networks

Zhengru Fang et al.

Summary: This paper discusses the EH-aided NGMA systems in B5G networks with MTCD devices and proposes sleep-scheduling policies to reduce energy consumption. By deriving closed-form expressions and proposing algorithms, the authors demonstrate that the proposed algorithms can reduce peak AoI at low power consumption across different multiple access protocols considered.

IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS (2022)

Article Geochemistry & Geophysics

Deep Learning for Unmanned Aerial Vehicle-Based Object Detection and Tracking: A Survey

Wu Xin et al.

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE (2022)

Article Computer Science, Artificial Intelligence

Detection and Tracking Meet Drones Challenge

Pengfei Zhu et al.

Summary: Drones equipped with cameras have been widely used in various fields, and the automatic understanding of visual data collected from drones has become highly demanding. To promote the development of object detection and tracking algorithms, challenge workshops have been organized, and a large-scale drone captured dataset, VisDrone, has been provided. The dataset enables extensive evaluation and investigation of visual analysis algorithms. The paper analyzes the current state of the field and proposes future directions.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2022)

Article Computer Science, Artificial Intelligence

Adaptive and stabilized real-time super-resolution control for UAV-assisted smart harbor surveillance platforms

Soyi Jung et al.

Summary: This study proposes a novel two-stage algorithm for industrial smart harbor and logistics platforms, utilizing drones for surveillance and introducing a scheduling algorithm for stability in the first stage.

JOURNAL OF REAL-TIME IMAGE PROCESSING (2021)

Article Computer Science, Information Systems

Seamless Dynamic Adaptive Streaming in LTE/Wi-Fi Integrated Network under Smartphone Resource Constraints

Jonghoe Koo et al.

IEEE TRANSACTIONS ON MOBILE COMPUTING (2019)

Proceedings Paper Computer Science, Artificial Intelligence

Minimum Delay Object Detection From Video

Dong Lao et al.

2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019) (2019)

Article Computer Science, Information Systems

A Survey of Deep Learning-Based Object Detection

Licheng Jiao et al.

IEEE ACCESS (2019)

Article Computer Science, Theory & Methods

A Light CNN for Deep Face Representation With Noisy Labels

Xiang Wu et al.

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY (2018)

Proceedings Paper Computer Science, Theory & Methods

Dynamic Security-Level Maximization for Stabilized Parallel Deep Learning Architectures in Surveillance Applications

Joongheon Kim et al.

2017 1ST IEEE SYMPOSIUM ON PRIVACY-AWARE COMPUTING (PAC) (2017)

Proceedings Paper Computer Science, Theory & Methods

REQUEST: Seamless Dynamic Adaptive Streaming over HTTP for Multi-Homed Smartphone under Resource Constraints

Jonghoe Koo et al.

PROCEEDINGS OF THE 2017 ACM MULTIMEDIA CONFERENCE (MM'17) (2017)

Article Computer Science, Hardware & Architecture

Quality-Aware Streaming and Scheduling for Device-to-Device Video Delivery

Joongheon Kim et al.

IEEE-ACM TRANSACTIONS ON NETWORKING (2016)

Article Computer Science, Information Systems

Energy optimal control for time-varying wireless networks

Michael J. Neely

IEEE TRANSACTIONS ON INFORMATION THEORY (2006)