Application of self-adaptive wavelet neural networks in ultrasonic detecting

Author

Yin, Xi-Peng ; Fan, Yang-yu ; Duan, Zhe-Min ; Cheng, Wei

Author_Institution

Dept. of Electron. Eng., Northwestern Polytech. Univ. (NPU), Xi´´an, China

fYear

2009

Firstpage

600

Lastpage

602

Abstract

It is important to remove the noise signal effectively in non-destructive testing. Using the wavelet and neural network algorithm, the author constructed self-adaptive wavelet neural networks in the ultrasonic testing. Better fitting signal is achieved by choosing Orthogonal Daubechies wavelet neuron and optimized scale parameter. The simulation results showed less distortion and better noise cancellation, and the method can be widely applied ton ultrasonic detecting.

Keywords

interference suppression; neural nets; self-adjusting systems; testing; ultrasonic applications; wavelet transforms; less distortion; neural network algorithm; noise cancellation; nondestructive testing; optimized scale parameter; orthogonal Daubechies wavelet neuron; self-adaptive wavelet neural network; ultrasonic detecting; ultrasonic testing; Automatic testing; Electronic equipment testing; Feedforward neural networks; Neural networks; Noise cancellation; Nondestructive testing; Signal analysis; Signal processing; Signal processing algorithms; Wavelet analysis; neural networks; self-adaptive; ultrasonic; wavelet analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Anti-counterfeiting, Security, and Identification in Communication, 2009. ASID 2009. 3rd International Conference on

Conference_Location

Hong Kong

Print_ISBN

978-1-4244-3883-9

Electronic_ISBN

978-1-4244-3884-6

Type

conf

DOI

10.1109/ICASID.2009.5276998

Filename

5276998