A novel IWPT-VMD denoising method for enhancing air-coupled ultrasonic detection of wood defects

To address the low signal-to-noise ratio in air-coupled ultrasonic detection of wood defects, this paper proposes a signal denoising method, improved threshold function-variational mode decomposition (IWPT-VMD), that integrates wavelet packet transform with an improved threshold function and variational mode decomposition. First, a continuous and adjustable improved threshold function is designed to effectively enhance the denoising performance of wavelet packet transform and significantly suppress the noise interference; second, the optimal number of modes in the variational mode decomposition of the preprocessed air-coupled ultrasound signal is determined, and the secondary processing is performed on the preprocessed signal in order to further eliminate the noise modes and optimize the signal quality. The experimental results show that the use of the IWPT-VMD method can effectively suppress the interference of background noise, and the signal-to-noise ratio is improved by about 25.62 dB compared with the original defect signal, which significantly enhances the contrast of the signal features between the defect region and the background region. In practical planar scanning imaging experiments, accurate quantification of defect areas is challenging with raw signal imaging. However, after signal processing using the IWPT-VMD method, the defect area detection accuracy in imaging improved to 88.1%. Experimental results validate the effectiveness and reliability of the proposed method in enhancing the precision of wood defect detection, providing robust technical support for non-destructive testing of wood.