采用双树复数小波变换对图像进行分解与重构,在BayesShrink阈值去噪的基础上,提出了基于小波系数层内和层间局域特性的自适应阈值去噪算法;构造出具有层内和层间局域特性的统计量和相应的映射,产生新的BayesShrink阈值.实验表明本方法能有效地去除图像中的白噪声,同时还能较好地保留图像的边缘信息,其效果优于目前的一些小波去噪方法. Based on BayesShrink threshold de-noising, an adaptive threshold de-noising algorithm based on local and inter-layer coefficients of wavelet coefficients was proposed. The algorithm was constructed based on double-tree complex wavelet transform. The statistics of local layer properties and corresponding mapping result in a new BayesShrink threshold.Experiments show that this method can effectively remove the white noise in the image and retain the edge information of the image better than the current Some wavelet denoising methods.