Nonlocal variational model and filter algorithm to remove multiplicative noise

摘要

Abstract. The nonlocal u0001NLu0002 means filter proposed by Buades, Coll, andnMorel u0001SIAM Multiscale Model. Simul. 4u00012u0002, 490–530, 2005u0002, whichnmakes full use of the redundancy information in images, has shown to benvery efficient for image denoising with Gauss noise added. On the basisnof the NL method and a striver to minimize the conditional mean-squarenerror, we design a NL means filter to remove multiplicative noise, andncombining the NL filter to regularity method, we propose a NL total varia-ntional u0001TVu0002 model and present a fast iterated algorithm for it. Experimentsndemonstrate that our algorithm is better than TV method; it is superior innpreserving small structures and textures and can obtain an improvementnin peak signal-to-noise ratio. © 2010 Society of Photo-Optical InstrumentationnEngineers. u0003DOI: 10.1117/1.3463019