Objects Detection of UAV for Anti-UAV Based on YOLOv4
Qingbang Shi, Jun Li
20202020 IEEE 2nd International Conference on Civil Aviation Safety and Information Technology (ICCASIT43 citationsDOI
Abstract
Currently, drone detection is a research hotspot in the security field due to its popularity. This paper proposed a recognition method of the low-altitude drone detection based on the YOLOv4 (You Only Look Once version 4) model and also introduced YOLOv4 algorithm in detection low-altitude UAV object for the first time. Sample set of drone flight attitude images constructed by shooting, downloading from the Internet and expanding the existing data is used to solve the lack of standard data set. The experimental results show that although YOLOv4, YOLOv3 and SSD algorithm all belong to one-stage algorithm, the average accuracy and real-time detection speed of YOLOv4 are better than that of the YOLOv3 and SSD.
Topics & Concepts
Computer scienceDroneArtificial intelligenceObject detectionUploadComputer visionPattern recognition (psychology)BiologyOperating systemGeneticsAdvanced Neural Network ApplicationsRobotics and Sensor-Based LocalizationVideo Surveillance and Tracking Methods