基于YOLOv8的无人机目标检测改进算法OA

With the rapid development of drone technology,the security risks it poses in daily activities and military fields are increasing day by day.In this context,it is urgent to study efficient detection algorithms targeting drones.A improved algorithm based on YOLOv8 is proposed to address the problem of difficulty in balancing the detection accuracy and computational complexity of small unmanned aerial vehicle targets in complex scenes using existing object detection algorithms.Firstly,the C2fRepGhost module is introduced into the backbone and neck net-works of YOLOv8 to reduce the computational complexity of the model;Secondly,introducing a small target detection head and eliminating large target detection heads and redundant networks to improve the recognition accuracy of the model for small targets;Finally,CBAM is intro-duced into the detection head structure to enhance the model's ability to pay attention to key features.The experimental results on the public dataset DUT Anti UAV show that the improved algorithm achieves an average accuracy of 95.3％while maintaining fast inference speed and low parameter count,which is 2.2％higher than YOLOv8m and reduces computational complexity by 69.2GFLOPs.