A Lightweight Uav Swarm Detection Method Integrated Attention Mechanism
نویسندگان
چکیده
Aiming at the problems of low detection accuracy and large computing resource consumption existing Unmanned Aerial Vehicle (UAV) algorithms for anti-UAV, this paper proposes a lightweight UAV swarm method based on You Only Look Once Version X (YOLOX). This uses depthwise separable convolution to simplify optimize network, greatly simplifies total parameters, while is only partially reduced. Meanwhile, Squeeze-and-Extraction (SE) module introduced into backbone improve model′s ability extract features; introduction Convolutional Block Attention Module (CBAM) in feature fusion network makes pay more attention important features suppress unnecessary features. Furthermore, Distance-IoU (DIoU) used replace Intersection over Union (IoU) calculate regression loss model optimization, data augmentation technology expand dataset achieve better effect. The experimental results show that mean Average Precision (mAP) proposed reaches 82.32%, approximately 2% higher than baseline model, number parameters about 1/10th YOLOX-S, with size 3.85 MB. approach is, thus, high suitable various edge devices.
منابع مشابه
Vision Based Obstacle Detection mechanism of a Fixed Wing UAV
In this paper we have developed a vision based navigation and obstacle detection mechanism for unmanned aerial vehicles (UAVs) which can be used effectively in GPS denied regions as well as in regions where remote controlled UAV navigation is impossible thus making the UAV more versatile and fully autonomous. We used a fixed single onboard video camera on the UAV that extracts images of the env...
متن کاملA Method for Forest Fire Detection Using UAV
This software provide functions on processing UAV (unmanned aerial vehicle) aerial image data according to the requirements of forestry area application on UAV platform. It gives a real-time and remote watch on fire in Greater Xing’ an Mountains, simultaneously the UAV is flying and getting the aerial data, helping users quickly master the number and location of fire points. Monitoring software...
متن کاملA Dynamic Path Generation Method for a UAV Swarm in the Urban Environment
In this paper, a dynamic path planning method for swarming of unmanned aerial vehicles (UAVs) in the urban environment is presented. For the missions of a team of UAVs in a complex environment, conflict-free paths for all vehicles should be calculated dynamically in real-time using the latest information on the changes in the surroundings such as pop-up threats. Therefore, we propose a hierarch...
متن کاملA New Particle Swarm Optimization Method for the Path Planning of UAV in 3D Environment PEHLIVANOGLU
Particle swarm optimization (PSO) method is relatively a new population-based intelligence algorithm and exhibits good performance in optimization problems. However, during the optimization process, the particles become more and more similar, and gather into the neighborhood of the best particle in the swarm, which makes the swarm prematurely converged possibly around the local solution. PSO te...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Drones
سال: 2022
ISSN: ['2504-446X']
DOI: https://doi.org/10.3390/drones7010013