Intelligent Recognition of Seismic Station Environmental Interference Based on YOLOv5

In recent years, human interference in seismic-station environments has posed challenges to the quality and accuracy of seismic signals, making data processing difficult. To accurately identify caused by personnel ensure reliability seismic-network instrument detection data, it is necessary track detected targets across consecutive frames. Deep neural networks have made significant progress this field. Therefore, an intelligent identification solution for environmental at stations proposed, which combines deep learning with multi-object tracking techniques. A centroid-matching algorithm based on Kalman filtering introduced entry/exit timestamps, alongside motion trajectories interfering individuals, thereby marking anomalous presence time-series data. Experimental results demonstrate that research provides effective station environments.