权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Debris-Flow Monitoring System with Acoustic Sensing and Digital Image Processing

具有声学传感和数字图像处理功能的泥石流监测系统

基本信息

批准号：
11680467
负责人：
ITAKURA Yasumasa
金额：
$ 2.18万
依托单位：
Shiga University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
1999
资助国家：
日本
起止时间：
1999 至 2001
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-11680467/
关键词：
debris-flow detection microphone sensor oil-immersed acoustic sensor improved hydrophone sensor computer-based spatial filtering velocimetry MPEG method spatio-temporal derivative space method cross-correlation method 改良型水中マイクロホン計算機対応空間

项目摘要

The purpose of this research is to develop a new monitoring system to prevent and to mitigate debris-flow hazards. The results of this research during three years are as follows:1. An acoustic sensor with a microphone inside a pipe set underground has a high signal-to-noise ratio compared with other ground vibration sensors, seismometer, piezoelectric sensor, or moving-coil sensor. An improved acoustic sensor having a microphone immersed in oil inside the pipe shows a possibility of twice detection range.2. A digital image processing based on video images of debris flow can make a proximity-detection by utilizing a detection algorithm of MPEG motion vectors. Another digital image processing methods such as computer-based spatial filtering velocimetry, spatio-temporal derivative space method, or cross-correlation method can estimate surface velocity of debris flow.3. A combination system of an acoustic sensor and an image-processing method may have a good performance of debris-flow monitoring. The acoustic sensor may compensate the image processing under a bad weather condition of deep fog or heavy rain : on the other hand, the image-processing method makes clear the real image of debris flow, which the acoustic sensor can't observe. Unfortunately, the feasibility test of this combination system has not been confirmed because debris flows have not occurred in a test torrent.4. A workshop on debris flow monitoring was held at the Swiss National Hydrological and Geological Survey in Ittigen-Bern, Switzerland, on 5-6 November 2001 by partial supporting of this found. The results of this research were presented in the workshop and took a good evaluation in viewing future monitoring of debris flow.

本研究的目的是开发一种新的监测系统，以预防和减轻泥石流灾害。本研究历时三年，主要研究结果如下：1.与其他地面振动传感器、地震仪、压电传感器或动圈传感器相比，在设置在地下的管道内具有麦克风的声学传感器具有高信噪比。一种改进的声传感器，其麦克风浸入管道内的油中，显示出两倍的检测范围的可能性。基于泥石流视频图像的数字图像处理，利用MPEG运动矢量检测算法进行邻近检测。另一种数字图像处理方法，如计算机空间滤波测速法、时空导数空间法、互相关法等可以估算泥石流表面速度.声传感器和图像处理方法的组合系统可以具有良好的泥石流监测性能。在浓雾或暴雨等恶劣天气条件下，声传感器可以补偿图像处理的不足：另一方面，该图像处理方法使声传感器无法观测到的真实的泥石流图像清晰化。遗憾的是，由于没有在试验洪流中发生泥石流，这种组合系统的可行性试验尚未得到证实。2001年11月5-6日，在瑞士伊蒂根-伯尔尼国家水文地质调查局的部分支持下，在瑞士伊蒂根-伯尔尼举办了一次泥石流监测研讨会，会上介绍了本研究的成果，并对今后的泥石流监测工作给予了很好的评价。