Recognition of UAVs in Infrared Images Based on YOLOv8
With the advancement of Unmanned Aerial Vehicle (UAV) technology, there has been a notable increase in the utilization of UAVs by criminals for engaging in illegal activities, which poses significant threats to critical infrastructure and national security. Previous studies have predominantly relied...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10755076/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536193009811456 |
|---|---|
| author | Gang Zhou Xiuqi Liu Hongliang Bi |
| author_facet | Gang Zhou Xiuqi Liu Hongliang Bi |
| author_sort | Gang Zhou |
| collection | DOAJ |
| description | With the advancement of Unmanned Aerial Vehicle (UAV) technology, there has been a notable increase in the utilization of UAVs by criminals for engaging in illegal activities, which poses significant threats to critical infrastructure and national security. Previous studies have predominantly relied on visible image recognition, which is vulnerable to variations in lighting conditions. Additionally, the performance of detection models is often hindered by the poor quality of images capturing UAV small targets. In contrast to visible light images, infrared images can capture the heat emitted by UAVs, rendering them less susceptible to fluctuations in lighting. To address the necessity for micro UAV detection in complex environments, this paper proposes an infrared image UAV detection method that employs super-resolution reconstruction through the Super-Resolution Convolutional Neural Network (SRCNN) and YOLOv8. Initially, a convolutional neural network is utilized to enhance the resolution and clarity of the infrared images. Subsequently, the YOLOv8 target detection model is applied to identify and detect UAVs. The results indicate that the average accuracy in detecting UAVs is 93.1%, which proves that it has excellent detection performance. |
| format | Article |
| id | doaj-art-96fa609f8c5e46a4b6070c66f0fc1a0e |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-96fa609f8c5e46a4b6070c66f0fc1a0e2025-01-15T00:01:53ZengIEEEIEEE Access2169-35362025-01-01131534154510.1109/ACCESS.2024.350058310755076Recognition of UAVs in Infrared Images Based on YOLOv8Gang Zhou0https://orcid.org/0000-0002-7448-6635Xiuqi Liu1Hongliang Bi2https://orcid.org/0009-0008-5094-6001School of Electrical Engineering, Naval University of Engineering, Wuhan, Hubei, ChinaSchool of Computer Science, Wuhan University, Wuhan, Hubei, ChinaSchool of Information and Control Engineering, China University of Mining and Technology, Xuzhou, Jiangsu, ChinaWith the advancement of Unmanned Aerial Vehicle (UAV) technology, there has been a notable increase in the utilization of UAVs by criminals for engaging in illegal activities, which poses significant threats to critical infrastructure and national security. Previous studies have predominantly relied on visible image recognition, which is vulnerable to variations in lighting conditions. Additionally, the performance of detection models is often hindered by the poor quality of images capturing UAV small targets. In contrast to visible light images, infrared images can capture the heat emitted by UAVs, rendering them less susceptible to fluctuations in lighting. To address the necessity for micro UAV detection in complex environments, this paper proposes an infrared image UAV detection method that employs super-resolution reconstruction through the Super-Resolution Convolutional Neural Network (SRCNN) and YOLOv8. Initially, a convolutional neural network is utilized to enhance the resolution and clarity of the infrared images. Subsequently, the YOLOv8 target detection model is applied to identify and detect UAVs. The results indicate that the average accuracy in detecting UAVs is 93.1%, which proves that it has excellent detection performance.https://ieeexplore.ieee.org/document/10755076/UAVinfrared imageSRCNNYOLOv8object detectionsuper-resolution |
| spellingShingle | Gang Zhou Xiuqi Liu Hongliang Bi Recognition of UAVs in Infrared Images Based on YOLOv8 IEEE Access UAV infrared image SRCNN YOLOv8 object detection super-resolution |
| title | Recognition of UAVs in Infrared Images Based on YOLOv8 |
| title_full | Recognition of UAVs in Infrared Images Based on YOLOv8 |
| title_fullStr | Recognition of UAVs in Infrared Images Based on YOLOv8 |
| title_full_unstemmed | Recognition of UAVs in Infrared Images Based on YOLOv8 |
| title_short | Recognition of UAVs in Infrared Images Based on YOLOv8 |
| title_sort | recognition of uavs in infrared images based on yolov8 |
| topic | UAV infrared image SRCNN YOLOv8 object detection super-resolution |
| url | https://ieeexplore.ieee.org/document/10755076/ |
| work_keys_str_mv | AT gangzhou recognitionofuavsininfraredimagesbasedonyolov8 AT xiuqiliu recognitionofuavsininfraredimagesbasedonyolov8 AT hongliangbi recognitionofuavsininfraredimagesbasedonyolov8 |