EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LRAPO-Based Adaptive Signal Imaging for Non-Contact Air-Coupled Ultrasonic Characterization of Natural Knots in Wood

Xinyi Zhao, Jun Wang

Forests · 14.11.2026 · 17 (11) · 1358

DOI: 10.3390/f17111358 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Forests, online erschienen am 14.11.2026. Verfasst von Xinyi Zhao und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Nanjing Forestry University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Natural knots are important wood quality features that alter local structure and can reduce the suitability of wood for processing and use. Air-coupled ultrasound offers non-contact testing, but the air–wood impedance mismatch and wood heterogeneity attenuate transmitted signals, making knot characterization difficult. This study proposes an adaptive signal-processing and imaging framework for the non-contact characterization of natural knots in wood. The framework introduces a Lifecycle Restructured Artificial Protozoa Optimizer (LRAPO) that adaptively determines the mode number and penalty factor of variational mode decomposition (VMD). After effective modes are selected and reconstructed, IQ demodulation is applied to extract a stable amplitude envelope for two-dimensional C-scan imaging. Simulations, ablation tests, robustness experiments, and weak-response tests show that LRAPO-driven VMD improves reconstruction accuracy and stability and preserves low-amplitude responses. At an input SNR of 9 dB, LRAPO-driven VMD achieved an output SNR of 23.0953 dB and an RMSE of 0.01846, corresponding to a 1.00 dB SNR gain and a 10.9% RMSE reduction compared with APO-driven VMD. Experiments on five wood specimens with natural knots show that the framework suppresses background fluctuations and clearly distinguishes knot regions from normal wood. LRAPO-driven VMD produced the highest CNR among the evaluated imaging strategies for all five specimens while preserving the lower envelope amplitudes associated with the knot regions. Relative to direct IQ-envelope imaging, the CNR increased by 28.9%–48.9% across the five specimens, with a mean improvement of 41.4%. The proposed approach provides a non-contact ultrasonic method for visualizing knot-related structural heterogeneity and supports wood quality assessment and nondestructive testing.

Förderung

  • Nanjing Forestry University

Zitieren

APA

Zhao, X., & Wang, J. (2026). LRAPO-Based Adaptive Signal Imaging for Non-Contact Air-Coupled Ultrasonic Characterization of Natural Knots in Wood. Forests, 17(11), 1358. https://doi.org/10.3390/f17111358

Vancouver

Zhao X, Wang J. LRAPO-Based Adaptive Signal Imaging for Non-Contact Air-Coupled Ultrasonic Characterization of Natural Knots in Wood. Forests. 2026;17(11):1358. doi:10.3390/f17111358

BibTeX
@article{Zhao2026,
  author = {Zhao, Xinyi and Wang, Jun},
  title = {LRAPO-Based Adaptive Signal Imaging for Non-Contact Air-Coupled Ultrasonic Characterization of Natural Knots in Wood},
  journal = {Forests},
  year = {2026},
  volume = {17},
  number = {11},
  pages = {1358},
  doi = {10.3390/f17111358}
}
Kostenlos starten →