Ultrasonic Signal Denoising Based on One-Dimensional Convolution and SWT-MRA Multi-Scale Representation
DOI:
https://doi.org/10.71204/6bsh1667Keywords:
Ultrasonic Signal Denoising, Stationary Wavelet Transform, Multiresolution Analysis, One-Dimensional Convolutional Neural Network, Multiscale RepresentationAbstract
Ultrasonic measurement signals are susceptible to random noise, impulsive artifacts, and high-frequency interference, while conventional wavelet thresholding may attenuate useful signal components. To address the trade-off between noise suppression and waveform preservation, this study proposes an ultrasonic signal denoising method that integrates stationary wavelet transform-based multiresolution analysis (SWT-MRA) with a one-dimensional convolutional neural network (CNN). An eight-level SWT-MRA decomposition using the sym8 wavelet produces one approximation component and eight equal-length detail components. A dilated residual CNN then learns a component-domain denoising mapping under joint waveform-, frequency-, gradient-, and component-level constraints. The method was evaluated on three independent experimental batches, and the learning-based models were repeatedly trained using five random seeds. On the independent test set, the proposed method achieved a mean absolute error (MAE) of 0.003685, a root mean square error (RMSE) of 0.005024, and an output signal-to-noise ratio (SNR) of 22.198 dB. Compared with the time-domain CNN, it reduced MAE and RMSE by 3.10% and 3.39%, respectively, while increasing output SNR by 0.297 dB. Compared with 1D U-Net, MAE and RMSE were reduced by 9.22% and 10.19%, respectively. Ablation experiments showed that the main performance gain was attributable to the explicit SWT-MRA multiscale representation, whereas component-level auxiliary supervision provided limited additional benefit. These results demonstrate that explicit multiscale representation can improve ultrasonic waveform recovery with only a modest increase in network size.
References
Abbate, A., Koay, J., Frankel, J., et al. (1997). Signal detection and noise suppression using a wavelet transform signal processor: Application to ultrasonic flaw detection. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 44(1), 14-26.
Cantero-Chinchilla, S., Wilcox, P. D., & Croxford, A. J. (2022). Deep learning in automated ultrasonic NDE—Developments, axioms and opportunities. NDT & E International, 131, 102703.
Cui, X., Li, D., Li, Z., et al. (2022). A GAN noise modeling based blind denoising method for guided waves. Measurement, 188, 110596.
Donoho, D. L. (1995). De-noising by soft-thresholding. IEEE Transactions on Information Theory, 41(3), 613-627.
Jadhav, S., Kuchibhotla, R., Agarwal, K., et al. (2023). Deep learning-based denoising of acoustic images generated with point contact method. Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems, 6(3), 031002.
Lázaro, J. C., San Emeterio, J. L., Ramos, A., et al. (2002). Influence of thresholding procedures in ultrasonic grain noise reduction using wavelets. Ultrasonics, 40(1-8), 263-267.
Li, X., Wang, Y., Wan, X., et al. (2023). Physics-informed deep filtering of ultrasonic guided waves for incipient defect inspection of large-scale square tube structures. Journal of Sound and Vibration, 567, 118066.
Loshchilov, I., & Hutter, F. (2019). Decoupled weight decay regularization. International Conference on Learning Representations.
Mallat, S. G. (1989). A theory for multiresolution signal decomposition: The wavelet representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 11(7), 674-693.
Nason, G. P., & Silverman, B. W. (1995). The stationary wavelet transform and some statistical applications. In A. Antoniadis & G. Oppenheim (Eds.), Wavelets and Statistics (pp. 281-300). Springer.
Pardo, E., San Emeterio, J. L., Rodriguez, M. A., et al. (2006). Noise reduction in ultrasonic NDT using undecimated wavelet transforms. Ultrasonics, 44(Suppl), e1063-e1067.
Qian, C., Xia, S., Zhang, P., et al. (2026). Application-specific guided-wave ultrasonic signal denoising: Knowledge-guided synthetic data pipeline and wavelet-initialized attention U-Net. Ultrasonics, 164, 108054.
Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015 (pp. 234-241). Springer.
Schmidt, P. L., Walker, D. G., Yuhas, D. J., et al. (2014). Thermal measurements using ultrasonic acoustical pyrometry. Ultrasonics, 54(4), 1029-1036.
Singh, H., Ahmed, A. S., Melandsø, F., et al. (2023). Ultrasonic image denoising using machine learning in point contact excitation and detection method. Ultrasonics, 127, 106834.
Tsai, W. Y., Chen, H. C., & Liao, T. L. (2006). High accuracy ultrasonic air temperature measurement using multi-frequency continuous wave. Sensors and Actuators A: Physical, 132(2), 526-532.
Wu, J., Yang, Y., Lin, Z., et al. (2025). Weak ultrasonic guided wave signal recognition based on one-dimensional convolutional neural network denoising autoencoder and its application to small defect detection in pipelines. Measurement, 242, 116234.
Zhang, Z., Du, X., Gao, F., et al. (2026). An improved U-Net method for denoising ultrasonic echo signals in carbon fiber composites. Ultrasonics, 157, 107782.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Weijiang Qiao, Yuhan Wang, Guanghan Ding, Yanlong Wei (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in this journal are licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are properly credited. Authors retain copyright of their work, and readers are free to copy, share, adapt, and build upon the material for any purpose, including commercial use, as long as appropriate attribution is given.
