Abstract
In this paper, an image de-noising method based on integer wavelet transform and Donoho's nonlinear wavelet thresholding is presented to post-process reconstructed image after compressing. If the image signal has abundant high-frequency details in wavelet domain, the estimation of threshold value in Donoho's de-noising algorithm will be interfered and some high-frequency information of the signal will be lost. So this paper incorporates adaptability into lifting algorithm to make distribution of image signal energy more concentrative, so that the threshold value can be depressed and then the PSNR of image is improved. The paper also combines redundant wavelet transform with adaptive wavelet transform to improve the de-noising performance of Donoho's thresholding algorithm.
