摘要
针对利用增雨弹进行人工增雨过程中,现有的基于单门限的端点检测算法的方法难以准确检测出声音信号端点,无法对增雨弹进行准确定位的缺陷。提出一种新的双阈值端点检测算法,首先利用巴特沃斯滤波器对声音信号进行去噪,初步去除干扰。然后利用声信号短时特征,设置端点检测阈值参数。最后,将4个感兴趣段进行提取,进而获得增雨弹炸点的准确位置。实验结果表明,所述方法能够有效提取感兴趣段的起始点与结束点,符合实验预期与后续实验要求,具有实际的应用价值。
In the process of artificial rain enhancement,the existing endpoint detection algorithm based on single threshold is difficult to detect the endpoint of sound signal accurately,and can not locate the rain enhancement accurately.In this paper,a new double threshold endpoint detection algorithm is proposed.Firstly,Butterworth filter is used to denoise the sound signal and remove the interference.Then,the threshold parameters of endpoint detection are set based on the short-term characteristics of acoustic signal.Finally,four interested segments are extracted to obtain the exact location of the detonation point.The experimental results show that the method described in this paper can effectively extract the starting point and the ending point of the interested segment,which is in line with the experimental expectations and the requirements of subsequent experiments,and has practical application value.
作者
姚丽娜
鲁旭涛
孙运强
柴晋强
Yao Lina;Lu Xutao;Sun Yunqiang;Chai Jinqiang(School of information and communication Engineering,North University of China,Taiyuan 030051,China)
出处
《国外电子测量技术》
2020年第1期6-9,共4页
Foreign Electronic Measurement Technology
基金
山西省应用基础研究项目(201701D221124)资助.
关键词
端点检测
巴特沃斯滤波
短时特征
双阈值
endpoint detection
Butterworth filtering
short-term features
double thresholds
作者简介
姚丽娜,硕士,主要研究方向为声定位、阵列信号处理。E-mail:1664478614@qq.com
