Abstract:This study applies four automatic seismic signal detection methods—short-long time window ratio,multi-band filtering,deep learning,and template matching filtering—to continuous waveform data recorded by the YUS,ZOD,and YUT stations,which are nearest to the epicenters of three major seismic events with well-documented foreshock activity. These events include the M7.1 Yushu earthquake in Qinghai on April 14,2010;the M5.9 earthquake at the junction of Shangri-La, Deqin,and Derong in Yunnan on August 31,2013;and the M7.3 Yutian earthquake in Xinjiang on February 12,2014. The study evaluates the effectiveness of each method in detecting foreshock signals and enhancing foreshock phase information. By comparing false negative and false positive rates,we propose using deep learning results to validate and supplement detections from the short-long time window ratio and multi-band filtering methods. These improved detections are then used as templates for matching and filtering. Additionally,the Generalized Phase Detection (GPD) algorithm is employed for the automatic identification of P and S phases,further enhancing the accuracy and completeness of foreshock phase identification.