A new image denoising algorithm based on adaptive threshold and fourth order partial diffusion equation

In this paper, a dual-tree complex wavelet transform (DTCWT) based hybrid image denoising algorithm which combines fourth order Partial Diffusion Equation (PDE) and adaptive thresholding is proposed for Gaussian noise corrupted images by considering the significant dependences of the wavelet coefficients across different scales. The DTCWT has the advantage of improved directional selectivity, approximate shift invariance, and perfect reconstruction over the discrete wavelet transform. The wavelet filter bank is used to decompose the image into approximation sub band and detail sub band. Though the noise affects both the sub bands the existing wavelet thresholding methods have the final noise reduced image with limited improvement. In the proposed algorithm the fourth order PDE technique is applied on the detail sub band and the adaptive thresholding is applied to the approximate sub band and tested on Gaussian noise corrupted images. The observation of visual quality and quantitative performance in terms of PSNR and SSIM shows improvement of the proposed method over the existing wavelet-based image denoising namely anisotropic diffusion, median filtering and diffusion equation techniques.