Aiming at the problem of de-noising signal blur caused by image information and noise aliasing in wavelet threshold de-noising algorithm

a threshold adaptive optimizing de-noising algorithm based on wavelet transform is proposed. On the one hand

considering the edge features of the image

the proportional shrinkage factor is introduced to separate the useful information of the image from the noise information. On the other hand

the Bayes Shrink threshold function is improved

and the optimal threshold of the noise image is adaptively searched by the PSO algorithm. The threshold function is improved for the difference of noise between wavelet decomposition layers

and adaptive de-noising is performed according to the distribution of noise in different directions and different scales. The de-noising experiment is performed on the one-dimensional signal and the twodimensional signal. Compared with the existing literature methods

the method can effectively remove the noise and better preserve the useful information in the original signal.