Suboptimality of nonlocal means on images with sharp edges

We conduct an asymptotic risk analysis of the nonlocal means image denoising algorithm for Horizon class images that are piecewise constant with a sharp edge discontinuity. We prove that the mean-square risk of nonlocal means is suboptimal and in fact is within a log factor of the mean square risk of wavelet thresholding.

[1]  A. Tsybakov,et al.  Minimax theory of image reconstruction , 1993 .

[2]  Saïd Ladjal,et al.  Exemplar-Based Inpainting from a Variational Point of View , 2010, SIAM J. Math. Anal..

[3]  Azriel Rosenfeld,et al.  Computer Vision , 1988, Adv. Comput..

[4]  David L. Donoho,et al.  De-noising by soft-thresholding , 1995, IEEE Trans. Inf. Theory.

[5]  D. Donoho Wedgelets: nearly minimax estimation of edges , 1999 .

[6]  Dana H. Ballard,et al.  Computer Vision , 1982 .

[7]  P. Bickel,et al.  Texture synthesis and nonparametric resampling of random fields , 2006, math/0611258.

[8]  S. Mallat A wavelet tour of signal processing , 1998 .

[9]  Pierrick Coupé,et al.  Bayesian Non-local Means Filter, Image Redundancy and Adaptive Dictionaries for Noise Removal , 2007, SSVM.

[10]  G. G. Stokes "J." , 1890, The New Yale Book of Quotations.

[11]  C. Stein Approximate computation of expectations , 1986 .

[12]  I. Johnstone,et al.  Minimax estimation via wavelet shrinkage , 1998 .

[13]  Jean-Michel Morel,et al.  A Review of Image Denoising Algorithms, with a New One , 2005, Multiscale Model. Simul..

[14]  E. Candès,et al.  Curvelets: A Surprisingly Effective Nonadaptive Representation for Objects with Edges , 2000 .

[15]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[16]  Sung Yong Shin,et al.  On pixel-based texture synthesis by non-parametric sampling , 2006, Comput. Graph..

[17]  John B. Shoven,et al.  I , Edinburgh Medical and Surgical Journal.

[18]  Jain Bimi,et al.  Digital Image Processing - An Introduction , 2013 .