Threshold-Varying Method of Stationary Wavelet Denoising for Ultrasonic Speckle Reduction

Abstract:

Ultrasonic inspection of the cast stainless steel with coarse grain size has not been possible to demonstrate unambiguously that current inspection techniques are adequate, which means it is very difficult for ultrasonic inspector to discriminate the reflected signals at the real flaw from scattered signals. In this paper, wavelet denoising is performed in the stationary wavelet domain by means of scale-varying threshold level to reduce the speckle and to enhance flaw-visibility. Experiments have been carried out on A-scan ultrasonic signals obtained from CCSS material which is a part of narrow gap welds connecting the steam generator to main reactor coolant piping in the nuclear power plant.

• Periodical: Materials Science Forum (Volumes 449-452)
• Pages: 1153-1156
• DOI: https://doi.org/10.4028/www.scientific.net/MSF.449-452.1153
• Online since: March 2004
• Authors: Chi Seong Park, Young Seock Lee, Seon Jin Kim
• Keywords: Speckle Noise, Ultrasonic Inspection, Wavelet Denoising

References

[1] D. L. Donoho: IEEE Trans. Inform. Theory, vol. 41 (1995), pp.613-627.

[2] N. M. Bilgutay and J. Saniie: Materials Evaluation, vol. 42 (1984), pp.808-814.

[3] A. Abbate, J. Koay, J. Frankel, S. C. Schroeder and P. Das: IEEE Trans. On UFFC, vol. 44, No. 1 (1997), pp.14-26.

[4] M. Lang, H. Guo and J. E. Odegard: IEEE Signal Processing Letters, vol. 3, No. 1 (1996).