Abstract: In order to enhance the signal to noise ratio (SNR) of fundamental ultrasonic echo signals for ultrasonic nondestructive testing (UNDT) and ultrasonic nondestructive evaluation (UNDE), an improved technique to suppress structural noises of ultrasonic signals on the basis of pattern recognition theory of support vector machine is presented. After the formation mechanism of structural noises is studied and the shortcomings of classical split spectrum processing (SSP) algorithm are analyzed, the fundamental ultrasonic signals are decomposed into wavelet domain by discrete wavelet transform. A signal and noise separator based on support vector machine(SVM) of which the kernel function is Gauss function is used to distinguish the target signals from the noises in wavelet domain, and the target signals are reconstructed to realize the aim of enhancing SNR by removing noises. The experimental results indicate that the presented technique has high performance reliability and can improve the SNR enhancing ability for ultrasonic target echo signals contaminated by structural noises compared with the classical SSP algorithm.

Key words: Signal to noise ratio (SNR), Support vector machine (SVM), Ultrasonic nondestructive testing(UNDT), Wavelet transform