Detection of humans in drone images using deep learning techniques
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Interest in creating reliable and fast human detection systems has increased due to the growing use of Unmanned Aerial Vehicles (UAVs) in a variety of applications, including search and rescue operations and surveillance. The YOLOv8 deep learning algorithm, specially designed for drone imagery, is used in this paper's real-time human recognition method. Because it is effective and precise at identifying people in complicated backdrops and with different lighting conditions, YOLOv8, an enhanced version of the You Only Look Once (YOLO) object detection algorithm, is used. The YOLOv8 model is trained on a large dataset of annotated drone images, encompassing diverse scenarios and environments to facilitate robust human detection. This study uses a UAV Human Dataset encompassing various environmental conditions, ensuring robust performance in challenging scenarios. This study contributes to advancing drone-based technologies for human detection, offering a scalable and efficient solution for real-world deployment. The YOLOv8 model achieved a precision of 88.83%, recall of 71%, and mAP score of 74%.
Fields of Study & Subjects
Advanced Neural Network Applications
Video Surveillance and Tracking Methods
Human Pose and Action Recognition
Publication details
- Journal / source
- Informatyka Automatyka Pomiary w Gospodarce i Ochronie Środowiska
- Volume
- 16
- Issue
- 1
- Pages
- 109–115
- Publication date
- 30 March 2026
- Language
- EN