Super-resolution using Hidden Markov model and Bayesian detection estimation framework

Abstract

This paper presents a new method for super-resolution (SR)reconstruction of a high-resolution (HR) image from severallow-resolution (LR) images. The HR image is assumed to be composedof homogeneous regions. Thus, the a priori distribution of thepixels is modeled by a finite mixture model (FMM) and a PottsMarkov model (PMM) for the labels. The whole a priori model isthen a hierarchical Markov model. The LR images are assumed to beobtained from the HR image by lowpass filtering, arbitrarilytranslation, decimation, and finally corruption by a random noise.The problem is then put in a Bayesian detection and estimationframework, and appropriate algorithms are developed based onMarkov chain Monte Carlo (MCMC) Gibbs sampling. At the end, wehave not only an estimate of the HR image but also an estimate ofthe classification labels which leads to a segmentation result.