基于音频识别的钱塘江潮涌实时监测技术

作者：吴佩贤1，王瑞荣1，王建中1，邹洪波1，薛安克1，孙映宏2，何晓洪2 日期：2009-04-21/span> 浏览：3635 查看PDF文档

基于音频识别的钱塘江潮涌实时监测技术

吴佩贤1，王瑞荣1，王建中1，邹洪波1，薛安克1，孙映宏2，何晓洪2
（1.杭州电子科技大学 信息与控制研究所,浙江 杭州 310018；2.杭州市水文水资源监测总站,浙江 杭州 310016）

摘要：钱塘江潮涌高且迅猛，伤人事故频发。为提高潮涌实时监测与预报水平，提出了一种基于音频能量幅值技术的潮涌识别方法。该方法通过采集沿江各危险点潮涌来临前后的声波，经滤波、快速傅里叶变换（FFT）处理后，对其作幅频特性分析，以提取潮涌音频能量幅值特征值，并进行自动识别。试验分析结果表明，该研究有效地解决了潮涌实时监测与预报。
关键词：钱塘江；潮涌；声音识别；自动识别；快速傅里叶变换
中图分类号：TP391.42文献标识码：A文章编号：1001-4551(2009)04-0074-04

Realtime monitoring technology on tidal water of Qiantang river based on audiorecognition
WU Peixian1, WANG Ruirong1, WANG Jianzhong1, ZOU Hongbo1,
XUE Anke1, SUN Yinghong2, HE Xiaohong2
(1. Institute of Information and Control, Hangzhou Dianzi University, Hangzhou 310018, China;
2. Hangzhou Hydrology and Water Resources Monitoring Station, Hangzhou 310016, China)
Abstract: Plenty of accidents have taken place in the tidal water of Qiantang river, which is high and fast. In order to enhance the realtime monitoring and forecasting, a new technology was presented. The technology was based on audio amplitude of the tidal identification method. By collecting the sound wave of the coming tidal water pre and post in every dangerous site along the river, through filtering and fast Fourier transform(FFT), amplitudefrequency characteristic analysis was made and the eigenvalue of audio amplitude of tidal water was extracted, then automatic identification was taken. The reseach results show that the method resolves realtime monitoring and forecasting efficaciously.
Key words: Qiantang river; tidal water; audiorecognition; automatic identification; fast Fourier transform(FFT)