Introduction - If you have any usage issues, please Google them yourself
Aiming at fault feature extraction in the background of strong noise, a wavelet denoising idea by incorporating neighboring coefficients is introduced into redundant second generation wavelet case. A denoising method of redundant second generation wavelet using neighbor dependency is presented. It overcomes the deficiency of traditional thresholding approaches. In this method, an original signal is decomposed by redundant second generation wavelet, and the detail and approximation signals have the same length as the original signal. The detail signals at each scale are then processed by neighbor dependency. Finally, the approximation signal and processed detail signals are reconstructed by inverse transform of redundant second generation wavelet to realize the signal denoising.