PROJECT TITLE :

Improvement Of Bm3d Algorithm Based On Wavelet And Directed Diffusion - 2017

ABSTRACT:

Block-matching and 3D filtering (BM3D) has shown nice power for image denoising. By decomposing the BM3D restored image into three orthogonal subbands, we tend to find that the subband of the high-frequency element, of that the wavelet coefficients are less than a given threshold, has made some true textures of the first image lost and, the PSNR of the restored image is pulled down. So, in this paper, we have a tendency to propose a modified BM3D algorithm to solve these issues. We diffuse the subband of the high-frequency element, of that the wavelet coefficients are less than the edge of the BM3D restored image to the corresponding subband of the noisy image to induce a replacement estimate by using an improved directed diffusion equation model. The Laplace operator is replaced by an anisotropic diffusion operator and 2 completely different coefficients are added within the 2 diffusion terms. By replacing the corresponding subband of the BM3D restored image with the new estimate, we have a tendency to get the improved denoising result. Experimental results demonstrate that the proposed technique achieves higher performance than original BM3D algorithm in terms of each removing noise and persevering the edges and details of the image.