地震信号中干扰噪声自动识别算法

Automatic Identification Algorithm for Interference Noise in Seismic Signals

中国地震台网中心地震观测台站数量不断增加,由其记录的地震数据快速增加,有效压制数据中的干扰噪声成为非常重要的工作。观测数据中干扰噪声位置不固定,传统的干扰噪声识别方法耗时长、精度低,难以满足日益增长的数据处理需求。为实现快速识别,精准处理观测数据中的干扰信号,文章提出一种决策树联合YOLOv5的噪声识别检测方法,实验数据表明,该方法对于干扰噪声识别具有很好的检测效果,干扰噪声识别精确率达到94.6%。

The number of seismic observation stations in the China Seismic Network Center is increasing,and the seismic data recorded by them is rapidly increasing,so it becomes very important to effectively suppress the interference noise in the data.The location of interference noise in observation data is not fixed,and the traditional interference noise identification method is time-consuming and low precision,which is difficult to meet the increasing demand of data processing.In order to achieve fast identification and accurate processing of interference signals in observation data,this paper proposes a noise identification detection method for decision tree joint YOLOv5,and the experimental data show that the method has good detection effect for interference noise identification,and the accuracy of interference noise identification reaches 94.6%.