Image Denoising Based on the Wavelet Semi-soft Threshold and Total Variation

Abstract

The wavelet threshold denoising method has some defects. For example, the hard threshold function has no continuity at the threshold, which causes the Gibbs ringing effect. The soft threshold is relatively smooth, but the image is blurred. Image denoising based on total variation (TV) can effectively preserve the edge detail of the image, but in the smooth area, the denoising effect is not good. In this paper, a total variation image denoising method based on the wavelet semi-soft threshold is proposed. First, the image is decomposed using the wavelet method and the semi-soft threshold method is used to denoise in the high layer. Then, the wavelet coefficients are used to reconstruct the image. The high-frequency components of the first layer are denoised using the total variation method. The wavelet coefficients of the layers reconstruct the image after denoising. The experimental results demonstrate that the proposed method has a higher PSNR (Peak signal to noise ratio) than other methods, and it can more effectively preserve image detail while the image is denoised.