1994
DOI: 10.1051/jp4:19945306
|View full text |Cite
|
Sign up to set email alerts
|

Analyse des échos ultrasonores par ondelettes

Abstract: We investigate the time-frequency localization of the continuous wavelet transform as a tool for extracting useful informations about targets from their ultrasonic echoes. A disbond problem is presented as an example. We also present and discuss the use of experimental ultrasonic echoes as analyzing wavelets. This technique is used to study the effect of targets geometry on the received echoes.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
2
0

Year Published

1996
1996
2000
2000

Publication Types

Select...
1
1

Relationship

1
1

Authors

Journals

citations
Cited by 2 publications
(2 citation statements)
references
References 3 publications
0
2
0
Order By: Relevance
“…Gaussian wavelets well satisfy this requirement because their Fourier transfOlID are also Gaussian, so they are invariant with FOUlier transfonn and have exponential decay. Another advantage is that such wavelets do not form an orthogonal basis, and it is well known that orthogonal wavelets are dyadic and require a dyadic subdivision of time and frequency axis [4]. Ultrasonic waves used in non destlUctive testing are wideband signals, but their bandwidth does not cover several octaves in frequency.…”
Section: Theory the Continuous Wavelet Transfonnmentioning
confidence: 99%
See 1 more Smart Citation
“…Gaussian wavelets well satisfy this requirement because their Fourier transfOlID are also Gaussian, so they are invariant with FOUlier transfonn and have exponential decay. Another advantage is that such wavelets do not form an orthogonal basis, and it is well known that orthogonal wavelets are dyadic and require a dyadic subdivision of time and frequency axis [4]. Ultrasonic waves used in non destlUctive testing are wideband signals, but their bandwidth does not cover several octaves in frequency.…”
Section: Theory the Continuous Wavelet Transfonnmentioning
confidence: 99%
“…The values of the parameters (a,b) are chosen as follows: a = 2 j + mlM land b = 1111, V max (6) where j is an octave resulting from the dyadic decomposition of the frequency axis, M is the number of voices considered in each octave and m is the CUlTent voice. The stability of the reconstruction algorithm requires at least 4 voices for each octave [4]. Then the resconstruction can be achieved by simply summing the wavelet coefficients.…”
Section: Theory the Continuous Wavelet Transfonnmentioning
confidence: 99%