High-Speed Drone Detection Based On Yolo-V8
Jun-Hwa Kim, Nam‐Ho Kim, Chee Sun Won
Abstract
Detecting drones in a video is a challenging problem due to their dynamic movements and varying range of scales. Moreover, since drone detection is often required for security, it should be as fast as possible. In this paper, we modify the state-of-the-art YOLO-V8 to achieve fast and reliable drone detection. Specifically, we add Multi-Scale Image Fusion and P2 Layer to the medium-size model (M-model) of YOLO-V8. Our model was evaluated in the 6th WOSDETC challenge.
Topics & Concepts
DroneComputer scienceArtificial intelligenceObject detectionComputer visionRange (aeronautics)Scale (ratio)Pattern recognition (psychology)EngineeringAerospace engineeringGeographyCartographyGeneticsBiologyAdvanced Neural Network ApplicationsVideo Surveillance and Tracking MethodsImage Enhancement Techniques