Application of ICAMM to Impact-Echo Testing - On Statistical Pattern Recognition in Independent Component Analysis Mixture Modelling - page 125

The theoretical and experimental demonstrations provided here allow the
application of the impact-echo method to be extended to the classification of
different kinds of defective materials. The knowledge of the condition of the
material can be enhanced, i.e., not only to detect whether a material is sound or
unsound, but to obtain greater knowledge about the defects inside the material.
The proposed procedure is intended to exploit to the maximum the information
obtained with the cost efficiency of only one single impact. There is a range of
future directions for this research, such as applying the proposed method in
industrial contexts and obtaining greater insights into the sources and mixing
matrices of the model in order to exploit all the information collected by the
sensors. From these insights, an accurate localization of the defects and a 3D
reconstruction of the internal structure of the material can be performed.
References
1. M. Sansalone, W.B. Streett, Impact-echo: Non-destructive Evaluation of Concrete and
Masonry (Bullbrier Press, USA, 1997)
2. N.J. Carino, The Impact-Echo Method: An Overview, ed by P.C Chang. Structures Congress
and Exposition (Washington D.C, 2001), pp. 1-18
3. A. Hyvärinen, J. Karhunen, E. Oja, Independent Component Analysis (John Wiley & Sons,
New York, 2001)
4. A. Cichocki, S. Amari, Adaptive Blind Signal and Image Processing: Learning algorithms
and applications (Wiley, John & Sons, New York, 2001)
5. C.F. Morabito, Independent component analysis and extraction techniques for NDT Data.
Materials Evaluation 58(1), 85-92 (2000)
6. R. Boustany, J. Antoni, Blind extraction of a cyclostationary signal using reduced-rank cyclic
regression - a unifying approach. Mech. Syst. Signal Process. 22, 520-541 (2008)
7. J. Antoni, Blind separation of vibration components: Principles and demonstrations. Mech.
Syst. Signal Process. 19, 1116-1180 (2005)
8. J.J. Gonzalez de la Rosa, R. Piotrkowski, J.E. Ruzzante, Higher order statistics and
independent component analysis for spectral characterization of acoustic emission signals in
steel pipes. IEEE Trans. Instrum. Meas. 56(6), 2312-232 (2007)
9. J.J. Gonzalez de la Rosa, C.G. Puntonet, I. Lloret, An application of the independent
component analysis to monitor acoustic emission signals generated by termite activity in
wood. Measurement 37, 63-76 (2005)
10. W. Li, F. Gu, A.D. Ball, A.Y.T. Leung, C.E. Phipps, A study of the noise from diesel engines
using the independent component analysis. Mech. Syst. Signal Process. 15(6), 1165-1184
(2001)
11. X. Liu, R.B. Randall, J. Antoni, Blind separation of internal combustion engine vibration
signals by a deflation method. Mech. Syst. Signal Process. 22, 1082-1091 (2008)
12. A. Salazar, L. Vergara, J. Igual, J. Gosalbez, R. Miralles, ICA model applied to multichannel
non-destructive evaluation by impact-echo. Lecture Notes in Computer Science, vol. 3195,
pp. 470-477 (2004)
13. A. Salazar, L. Vergara, J. Igual, J. Gosalbez, Blind source separation for classification and
detection of flaws in impact-echo testing. Mech. Syst. Signal Process. 19(6), 1312-1325
(2005)
14. L.K. Hansen, M. Dyrholm, A prediction matrix approach to convolutive ICA, in Proceedings
of IEEE Workshop on Neural Networks for Signal Processing XIII, pp. 249-258 (2003)