超声检测信号直达波衰减的Curvelet变换方法  

Direct wave attenuation technique of ultrasonic detection signal by Curvelet transform

作　　者：王凡[1,2] 肖建平[1] 朱自强[1] 密士文[1] 

机构地区：[1]中南大学地球科学与信息物理学院,长沙410083 [2]中国水电顾问集团贵阳勘察设计研究院,贵阳550015

出　　处：《物探化探计算技术》2014年第1期22-27,共6页Computing Techniques For Geophysical and Geochemical Exploration

基　　金：国家自然科学基金项目(41274122);湖南省自然科学基金项目(12JJ6033);中南大学中央高校基本科研业务费专项资金项目(2011QNZT019)

摘　　要：利用超声脉冲回波法检测混凝土内部缺陷体时,由于介质的各向异性,使得接收信号中含有较强的结构噪声;同时,直达波信号振幅强,衰减少,传播路径短。上述原因导致难以分辨测量信号中的有效信号。为了压制噪声,去除直达波,采用了Curvelet变换对实测超声数据进行变换处理,分析直达波在Curvelet域的分布特点,得到其Curvelet系数,采用类似去噪的手段选取合适的阈值,重构得到滤除直达波的有效信号。同小波变换相比,Curvelet变换在去除直达波、滤除噪声的同时,较好地保留了有效信号。对比经过Curvelet变换处理前、后数据的SAFT成像效果表明,Curvelet变换能极大地提高超声无损检测数据分析的准确性。Using ultrasonic pulse echo method to detect internal defects in concrete body,because of the medium anisotropy, making the receiving signal contains strong structure noise,at the same time,the direct wave signal amplitude is strong,and it hardly to reduce,the pathway is also very short.For these reasons make it difficult to distinguish the effective signal between a valid signal.In order to suppress noise and remove the direct wave.Curvelet transform have been used to analysis the measured ultrasonic data in this article.By analysis of the direct wave in the distribution characteristics in Curvelet domain,and get the Curvelet coefficient.Using a way similar to denoise to reconstruction the signal after filter the direct wave.Compared with the wavelet transform,the Curvelet transform can remove direct wave,filtering noise and also good at retaining the effective signal.Using the data analysis by Curvelet transform to get a SAFT image indicate that Curvelet transform can greatly improve the accuracy of the data analysis in the ultrasonic nondestructive testing.

关 键 词：超声波 CURVELET变换 直达波衰减 去噪 

分 类 号：P631.5[天文地球—地质矿产勘探]