An eigenvalue-based approach is proposed for enhancement of eddy current images. The
images are obtained by raster scanning an absolute eddy current probe over defects in a ferromagnetic
carbon steel plate and by measuring the changes in probe impedance. During
eddy current testing of ferromagnetic materials, variations in local magnetic permeability
produce incoherent noise as compared to systematic impedance changes due to defects.
In the proposed approach, eigen pairs of image covariance matrix are determined, and only
the significant eigen pairs that have maximum information of the impedance changes due
to defects are utilized to reconstruct images free from noise attributable to permeability
variations. The capability of the proposed approach is assessed using eddy current images
of a hole and a machined notch in a carbon steel plate.
1. S. S. Udpa and P. O. Moore. Nondestructive Testing Handbook, Electromagnetic Testing, 2nd ed., Vol.
5. Columbus, OH, ASNT (2004).
2. V. S. Cecco, G. Van Drunen, and F. L. Sharp. Eddy Current Manual, AECL Report AECL-7523, Vol. 1,
Chalk River Laboratories, Ontario, Canada (1983).
3. S. Thirunavukkarasu, S. Vaidyanathan, B. P. C. Rao, K. H. Mahendran, T. Jayakumar, P. Kalyanasundaram,
and B. Raj. Development of eddy current non-destructive testing procedure for quality control
of thin walled nickel tubes. Insight 46(9):533–536 (2004).
4. E.-L. Kim, S.-J. Song, Y. H. Kim, C.-J. Yim, J.-H. Park, K.-B. Kim, and Y. H. Choi. Analysis of eddy current
testing signals for quantitative flaw characterization in steam generator tubes using principle
component analysis. Proceedings of the 16th World Conference on Nondestructive Testing, Montreal,
Canada (2004). www.ndt.net/article/wcndt 2004/index.htm
5. A. Sophian, G. Y. Tian, D. Taylor, and J. Rudlin. A feature extraction technique based on principal
component analysis for pulsed eddy current NDT. NDT&E International 36(1):37–41 (2003).
6. J. C. Nash. The Algebraic Eigen Value Problem. In Compact Numerical Methods for Computers:
Linear Algebra and Function Minimisation, 2nd ed., Chapter 9, pp. 102–118 (1990). Adam Hilger, Bristol,
7. J. Jiao, D. Negut, and B. Link. Multiple attenuation using eigen value decomposition. Proceedings of
the 66th Annual International Meeting, Vol. 18, pp. 1052–1055 (1999). Society of Exploration
Geophysicists, Tulsa, Oklahoma.
8. S. Thirunavukkarasu, B. P. C. Rao, T. Jayakumar, P. Kalyanasundaram, and B. Raj. Quantitative eddy
current testing using radial basis function neural networks. Materials Evaluation 62(12):1213–1217
9. B. P. C. Rao, B. Raj, T. Jayakumar, and P. Kalyanasundaram. An artificial neural network for eddy
current testing of austenitic stainless steel welds. NDT&E International 35(6):393–398 (2002).