Have a personal or library account? Click to login

Lightweight Low-Altitude UAV Object Detection Based on Improved YOLOv5s

By:
Open Access
|Mar 2024

Figures & Tables

Figure 1.

ATD-YOLO Network Structure
ATD-YOLO Network Structure

Figure 2.

Framework of image measuring system [14]
Framework of image measuring system [14]

Figure 3.

C3F structural schematic diagram
C3F structural schematic diagram

Figure 4.

CARAFE upsampling calculation flowchart
CARAFE upsampling calculation flowchart

Figure 5.

EMA Attention Mechanism
EMA Attention Mechanism

Figure 6.

GSConv Module
GSConv Module

Figure 7.

VoVGCSP Module
VoVGCSP Module

Figure 8.

The positions of GSConv and VOVGCSP modules
The positions of GSConv and VOVGCSP modules

Figure 9.

Length and Width Distribution Chart of the Anti-Mini Drone Dataset
Length and Width Distribution Chart of the Anti-Mini Drone Dataset

Figure 10.

Samples of simple background from Anti-Mini Drone
Samples of simple background from Anti-Mini Drone

Figure 11.

Samples of complex background from Anti-Mini Drone
Samples of complex background from Anti-Mini Drone

Figure 12.

PR curves for various feature extraction modules (IOU=0.5)
PR curves for various feature extraction modules (IOU=0.5)

Figure 13.

PR curves for various feature extraction modules (IOU=0.5)
PR curves for various feature extraction modules (IOU=0.5)

Figure 14.

Model PR curve (IOU=0.5)
Model PR curve (IOU=0.5)

Figure 15.

PR curves of mainstream algorithms on the test set (IOU=0.5)
PR curves of mainstream algorithms on the test set (IOU=0.5)

Figure 16.

Object detection outcomes in diverse scenarios
Object detection outcomes in diverse scenarios

Figure 17.

Object detection outcomes in a consistent scenario
Object detection outcomes in a consistent scenario

Mainstream Algorithm Comparative Experiment Results

ModuleParams/106GFLOP/GAP.5/%FPS
YOLOv3 Tiny8.6612.979.1166.67
YOLOv5s7.0115.992.268.79
YOLOv7 Tiny6.0113.288.463.30
YOLOv8s11.1228.489.0109.89
ATD-YOLO5.2311.092.875.35

Results of ablation experiments

YOLOv5sC3FEMACARFESlim-NeckParams/10 6GFLOP/GmAP.5/%FPS
7.0115.892.268.79
6.3313.892.375.05
6.3814.192.769.83
6.4014.193.167.85
5.2311.092.875.35

Contrast experiment of attention module

ModulemAP.5/%GFLOP /GParams/106FPS
SE[24]91.813.86.3774.93
ECA[25]92.213.86.3474.37
CBAM[26]92.313.86.3772.63
CA[27]91.213.86.3672.87
EMA92.714.16.3869.83

Origin of the Dataset and Quantity of Images

DatasetNumber of Images
Det-Fly3893
Drone-vs-Bird3959
Real World1525
Multi-view drone tracking3447
DUT anti-UAV3639
Anti-UAV2767

Experimental Setup Configuration

NameEnvironment Configuration
System EnvironmentUbuntu 22.04
CPUAMD Ryzen 9 5950X
GPURTX 4060 Ti 16GB
Deep Learning FrameworkPytorch 1.13.1
IDECUDA 11.7
Language: English
Page range: 87 - 99
Published on: Mar 28, 2024
Published by: Sciendo
In partnership with: Paradigm Publishing Services
Publication frequency: 4 times per year

© 2024 Haokai Zeng, Jing Li, Liping Qu, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.