Robust and computationally efficient superresolution algorithm

Abstract

Superresolution is the process of combining information from multiple subpixel-shifted low-resolution images to form a high-resolution image. It works quite well under ideal conditions but deteriorates rapidly with inaccuracies in motion estimates. We model the original high-resolution image as a Markov random field (MRF) with a discontinuity adaptive regularizer. Given the low-resolution observations, an estimate of the superresolved image is obtained by using the iterated conditional modes (ICM) algorithm, which maximizes the local posterior conditional probability sequentially. The proposed method not only preserves edges but also lends robustness to errors in the estimates of motion and blur parameters. We derive theoretically the neighborhood structure for the posterior distribution in the presence of warping, blurring, and downsampling operations and use this to effectively reduce the overall computations. Results are given on synthetic as well as real data to validate our method.

Image Quality Assessment by Upsampling Methods Based on Spatial Spectrum Extrapolation

“…One of the most popular Bayesian-based SR approaches is the maximum a posteriori (MAP) estimation approach [27,28,29,30,31,32]. The MAP estimator of x maximizes the a posteriori PDF P (x|y k ) with respect to x, i.e.…”

A survey on super-resolution imaging

“…In general, the SR image techniques can be classified into four classes: (i) frequencydomain-based approach [21,[38][39][40][41][42][43], (ii) interpolation-based approach [44][45][46][47], (iii) regularization-based approach [48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66], and (iv) learning-based approach [67][68][69][70][71][72][73][74]. The first three categories get a higher-resolution image from a set of lowerresolution input images, while the last one achieves the same objective by exploiting the information provided by an image database.…”

“…Motivated by the fact that the SR computation is, in essence, an ill-posed inverse problem [78], numerous regularizationbased SR algorithms have been developed for addressing this issue [49,50,52,[54][55][56]. The basic idea of these regularization-based SR approaches is to use the regularization strategy to incorporate the prior knowledge of the unknown high-resolution image.…”

“…Two most popular Bayesian-based SR approaches are maximum likelihood (ML) estimation approach [48] and maximum a posterior (MAP) estimation approach [49,50,52,[54][55][56].…”

Image Interpolation Based on Saliency Information

“…To try to get better performance, some studies propose image super-resolution algorithms based on regularization [15][16][17]. The basic idea of these methods is to add priori knowledge of high resolution image.…”