Low Altitude Small UAV Detection Based on YOLO model

Abstract

Low altitude small UAV recently becomes a hot technology with rapid development and comprehensive application which raising many risks in Low-altitude airspace. Considering the characteristics of low altitude small UAV, previous approaches like radar and radio are insufficient, this paper is devoted to the detection of small UAV by using images from low-cost cameras. Method of detecting low altitude small UAV based on YOLO model with two neural networks, ResNet and DenseNet, is designed and performed. One small dataset is established accordingly. Experiment on the dataset shows that the detection method with YOLO model can make contributions to the low altitude UAV detection in complicated environments.