超声缺陷检测中小波变换信号处理的参数确定  

Wavelet transform signal processing: determination of the processing parameters in ultrasonic flaw detection

作　　者：张广明[1] 赵明涛[2] 王裕文[2] 谭玉山[2] 

机构地区：[1]南京大学,南京210093 [2]西安交通大学,西安710049

出　　处：《声学学报》2000年第5期450-454,共5页Acta Acustica

摘　　要：小波变换是最近的处理时变信号的技术，Ａ．Ａｂｂａｔｅ，Ｊ．Ｋｏａｙ和Ｊ．Ｆｒａｎｋｅｌ等提出了一种基于连续小波变换的小波信号处理器进行超声信号处理。但是在应用小波信号处理器进行超声信号处理时，未考虑处理参数的影响，没有进行参数优化。本文从时频局部化角度分析了超声缺陷检测中小波信号处理器的参数优化问题，给出了各种参数如：母小波、母小波的频率带宽比和滤波器数目等的选择方法，并进行了实验验证。The wavelet transform is the most recently technique for processing signals with time-varying spectra. A wavelet transform signal processor based on continuous wavelet transform has been proposed by several authors for processing ultrasonic signals. In this paper, wavelet transform signal processor parameter optimizing in ultrasonic flaw detection is studied as viewed from time-frequency localization. Additionally, the determinate methods of the parameters, such as the mother wavelet, the frequency-to-bandwidth ratio of the mother wavelet , the number of filters and etc., are also given. Finally, the experiment verification is performed.

关 键 词：超声 缺陷检测 小波变换 信号处理 参数 

分 类 号：O426[理学—声学]