基于智能视觉检测与语义分析的反无人机实时预警系统OA

To address the security threats from unauthorized and erratic UAV flights in low-altitude airspace de-fense and the technical challenge of achieving real-time and precise early warning in complex settings,this paper proposes a real-time UAV intrusion warning system that integrates the YOLOv10 object detection algorithm with the multimodal large language model of General Language Model-4V(GLM-4V).The system hardware consists of high-sensitivity electro-optical detection equipment and a central computing unit,while the software integrates a client in-teraction platform and a multimodal data processing module.At the algorithmic level,YOLOv10 is employed for rap-id UAV localization,supplemented by the SORT algorithm for dynamic trajectory tracking.An innovative framework for semantic description generation based on GLM-4V is designed,which incorporates a structured prompt template to convert detection results(target location,attributes,and environmental background)into compact textual descrip-tions in real time.This significantly alleviates the communication burden in bandwidth-limited scenarios.Experimen-tal results demonstrate that under extreme conditions—maximum detection range of 1 km and maximum flight speed of 23 m/s—the system achieves an identification accuracy of 97.6%.By forming a closed-loop process of visual de-tection-semantic analysis-dynamic transmission,the proposed system effectively tackles the challenge of real-time early warning against UAV intrusions in complex environments,thereby offering a highly robust solution for low-alti-tude airspace security.