Anti-Drone System: A Visual-based Drone Detection using Neural Networks
Ann Janeth Garcia, Jae Min Lee, Dong‐Seong Kim
Abstract
A system that secures an area from trespassing drones that might bring threat is in demand these days since drones became easily-available to the public and it became easier to operate. This paper proposes an anti-drone system that uses visual sensing to detect drones. A Faster R-CNN (Region-based Convolutional Neural Network) with ResNet-101 (Residual Neural Network-101) networks are used in this paper using a dataset from the SafeShore project. The network's accuracy is 93.40% and it has successfully detected drones in the simulation that has been done.
Topics & Concepts
DroneConvolutional neural networkComputer scienceResidualArtificial intelligenceArtificial neural networkResidual neural networkComputer visionAlgorithmBiologyGeneticsVideo Surveillance and Tracking MethodsAdvanced Neural Network ApplicationsFire Detection and Safety Systems