自适应小波阈值函数在图像增强中的应用研究OA

Conventional wavelet thresholding algorithms exhibit issues of non-smoothness and discontinuity at the threshold position during wavelet transformation,leading to the loss of image details and poor enhancement effects when processing noisy images.In view of this,a self-adaptive wavelet threshold function based on the softsign(x)function is reconstructed.This function alleviates the gradient vanishing effectively.In addition,a contraction factor is introduced,and self-adaption adjustment is carried out according to the number of wavelet decomposition layers to distinguish between useful information and noise accurately,thereby enhancing the image enhancement effect.Comparisons were made between conventional threshold functions and other improved threshold functions by simulation experiments.The results indicate that the self-adaptive wavelet threshold function proposed in this study has a significant effect in denoising and enhancing image details,and can effectively enhance the edge and texture information of noisy images,outperforming the other methods.