What is it about?
Convolutional Neural Networks (CNNs) have become popular for object detection, particularly in identifying unmanned aerial vehicles (UAVs). This research focuses on identifying UAVs using video footage captured simultaneously by a visible light camera and an infrared camera, resulting in difficult viewing angles and significant motion blur. Two models, You Only Look Once (YOLO) and Single Shot MultiBox Detector (SSD), were evaluated for their robustness in detecting UAVs under motion blur conditions.
Featured Image
Photo by Jordan Cormack on Unsplash
Why is it important?
Novel architecture by modified YOLO and SSD models improves feature extraction and detection accuracy, particularly for small targets and motion blur which are often challenging to identify in UAV imagery.
Perspectives
Deep Convolutional Neural Networks (CNNs) have emerged as a pivotal technology for detecting Unmanned Aerial Vehicles (UAVs) due to their ability to process and analyze visual data effectively. Our studies demonstrated that modified models (Yolo & SSD) have a good ability to tickle the challenges in UAV detection, which include their small size, resemblance to birds, and varying environmental conditions, which complicate recognition tasks.
MUTAGISHA NORBELT
Beihang University
Read the Original
This page is a summary of: Detecting UAV Target with Deep Convolutional Neural Network, July 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3686625.3686628.
You can read the full text:
Contributors
The following have contributed to this page
