Application of change detection technique to ultrasound signal processing. Use of spectral parametric estimation, time-frequency and wavelet

Concerns the use of parametric-adaptive tools for ultrasound signal processing in nondestructive materials testing. A wave generated by a piezoelectric transducer is propagated across a sample, and the response signal is processed. The spectral ratio method used for calculating the quality factor of a material is not very accurate when Fourier analysis is used. We show that determination of the quality factor (or attenuation factor of a material) can be obtained using parametric spectral estimation techniques rather than nonparametric ones (Fourier). By use of abrupt change detection, we determine the signal length to be used in spectral ratio method without loss of information. This can be verified also by use of time-frequency, time-scale, and segmentation.