基于改进LOF的次声事件检测算法

his paper proposes a new infrasound event detection algorithm, FF-LOF, to address the issues of high false detection and poor robustness rate in the widely used PMCC algorithm. The algorithm is based on the idea of outlier detection in machine learning. It processes multi-channel waveform data by dividing frames, extracts multiple correlation features of each frame, and uses the principal component analysis algorithm for feature fusion. Then, the LOF algorithm is used to detect outliers within the extracted feature values of each frame, thereby determining the range of infrasound events. Real infrasound data collected from monitoring stations is used as the background for evaluation. The detection results of FF-LOF algorithm are compared with those of the PMCC algorithm under multiple rounds of Bayesian optimization. The experimental results demonstrate that the proposed algorithm exhibits higher robustness and lower false detection rate.