Markov Chain Monte Carlo Super-resolution Image Reconstruction With Artifacts Suppression
Jing Tian, Kai‐Kuang Ma · 2006
Recently, the Markov chain Monte Carlo (MCMC) technique has been proved as an effective approach to address the super-resolution image reconstruction problem. However, this approach usually requires a substantial amount of time to generate a sufficiently large number of samples for estimating the unknown high-resolution image. Limiting the simulation time could possibly lead to some artifacts presented in the reconstructed high-resolution image. To effectively mitigate these artifacts, an outlier-sensitive bilateral filtering is proposed in this paper, which contains a switching mechanism steered by an outliers detection scheme. Only for those pixel positions that have been identified containing outliers, our proposed bilateral filtering will be applied; for the rest, the conventional bilateral filtering will be exploited. Experimental results are presented to demonstrate the superior performance of the proposed method