11th Annual International Conference on Industrial Engineering and Operations Management

Establishing Drone Detection System By Using Deep Learning And YOLOv3 Formatting

Faisal AlMeshkhas, AbdulAziz Bin Shuhaywin & Sobhi Mejjaouli
Publisher: IEOM Society International
0 Paper Citations
1 Views
1 Downloads
Track: Artificial Intelligence
Abstract

Using a machine-vision detection based on deep learning system, the study established using darknet framework to identify and making the drone detection system. Using the darknet, YOLOv3 algorithm and OpenCV, the system was implemented on our computer to identify drones based on the live feed obtained from camera or uploaded image. The system was implemented to track and recognize drones on a different background. The project methodology is based on four different stages, gathering the dataset, annotation, training the system and testing the system. The project was able to identify the drones with an average success rate of 99% and 100% from live videos and uploaded images, respectively. The project showed that because of the deep learning strength, the detection was robust against changes in light intensity and the existence of background effects in different environmental conditions.

Published in: 11th Annual International Conference on Industrial Engineering and Operations Management, Singapore, Singapore

Publisher: IEOM Society International
Date of Conference: March 7-11, 2021

ISBN: 978-1-7923-6124-1
ISSN/E-ISSN: 2169-8767