5.Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering
K.Dabov,A.Foi,V.Katkovnik and K.Egiazarian, IEEE Transactions on Image Processing, vol.16, no.8, August 2007
The denoising performance of BM3D algorithm shows good preservation of uniform areas, smooth intensity transitions, textures, repeating patterns,sharp edges and singularities. In Particular, the repeated patterns are effectively reconstructed. With respect to visual quality, the image details are well preserved and very few artifacts are introduced.This method achieves good denoising performance in terms of both Peak signal-to-noise ratio(PSNR) and subjective visual quality.
Software:http://www.cs.tut.fi/~foi/GCF-BM3D/index.html#ref_software
Results for MATLAB Code
Basic Estimate, PSNR :31.37 dB
Final Estimate: (total time : 4.7 sec), PSNR: 32.08 dB
ans= 32.0775