基于小波自适应阈值去噪的MT信号处理方法  

Method of processing magnetotelluric signal based on the adaptive threshold wavelet

作　　者：蔡剑华[1] 肖晓[2] 

机构地区：[1]湖南文理学院信息研究所,常德415000 [2]中南大学地球科学与信息物理学院,长沙41000

出　　处：《地球物理学进展》2015年第6期2433-2439,共7页Progress in Geophysics

基　　金：国家自然科学基金项目(41304098);湖南省自然科学基金项目(12JJ4034);湖南省教育厅青年项目(13B076);湖南省重点建设学科-光学基金;湖南省重点实验室"光电信息集成与光学制造技术"和"湖南省光电信息技术校企联合人才培养基地"联合资助

摘　　要：针对常用大地电磁(magnetotelluric,MT)信号小波阈值去噪中阈值函数和阈值选取的不足,选取一种新的阈值函数,并提出了基于小波变换多分辨Stein无偏风险估计,自适应的获得最优阈值的方法;给出了新函数和自适应阈值的确定方法;在与小波硬、软阈值去噪、小波模极大值法去噪结果对比的基础上,用仿真实验验证了方法的可行性;最后用内蒙某地实测的大地电磁数据进行了去噪效果的对比研究.结果表明:本文选取的新函数和自适应阈值的确定方法是正确、有效的,克服了常用小波硬、软阈值去噪的缺点;去噪后大地电磁信号变得平稳,估算的响应参数方差减小,曲线更为圆滑、连续,为后续的地质解释和评价提供了更为准确的信息.In view of the deficiency from the wavelet threshold method commonly used in magnetotelluric signals procession,a new threshold function is given and the threshold is obtained adaptively based on Stein unbiased risk estimation in multiresolution of wavelet transform.The determination method of new function and adaptive threshold is presented.Based on the comparison with wavelet hard threshold,soft threshold method and module maxima method,the proposed method is verified to be reliable from the simulation results.Finally some measured data from Inner Mongolia is processed similarly.The results show that the proposed method to determine the threshold function and the adaptive threshold is correct and effective.The new method overcomes the disadvantage from the traditional wavelet threshold method.After de-noising,magnetotelluric signals become stable, the variance estimation response parameters is reduced,and the parameter curve is more smooth and continuous.The better de-noised results provide more accurate information for subsequent geological interpretation and evaluation.

关 键 词：大地电磁信号 阈值函数 自适应 小波阈值 去噪 

分 类 号：P631[天文地球—地质矿产勘探]