Nonlocal‐means single‐frame image zooming

Abstract

In this paper we present a novel single‐frame image zooming technique which is inspired by fractal‐based image zooming, example‐based zooming, and nonlocal‐means image denoising and combines these techniques in a consistent and improved framework. In Bayesian terms, this example‐based zooming technique targets the minimum mean square error (MMSE) estimate by learning the posterior directly from examples taken from the image itself at a different scale, similar to fractalbased techniques. The examples are weighted according to a scheme introduced by Buades et al. to perform nonlocal‐means image denoising. Finally, various computational issues are addressed and some results of this image zooming method applied to natural images are presented. (© 2008 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)