A Unified Approach to Superresolution and Multichannel Blind Deconvolution

Abstract

This paper presents a new approach to the blind deconvolution and superresolution problem of multiple degraded low-resolution frames of the original scene. We do not assume any prior information about the shape of degradation blurs. The proposed approach consists of building a regularized energy function and minimizing it with respect to the original image and blurs, where regularization is carried out in both the image and blur domains. The image regularization based on variational principles maintains stable performance under severe noise corruption. The blur regularization guarantees consistency of the solution by exploiting differences among the acquired low-resolution images. Several experiments on synthetic and real data illustrate the robustness and utilization of the proposed technique in real applications.

Image Deblurring via Alternating Direction Method of Multipliers

“…Min ||A*w*x-y +Tф (x) (12) This technique introduces a multi frame super resolution (MFSR) to estimate high resolution image from a low resolution images [8]. This MSFR problem can be reformulated into a problem of multi frame blind deblurring (MFDB).This technique adopt a matrix vector notation.…”

Generic Feasibility of Perfect Reconstruction With Short FIR Filters in Multichannel Systems

“…For this class of problems only the nonsubsampled case, i.e., the problem of blind multi-channel deconvolution, is fully studied (see [39], [40] and references therein). In subsampled systems little rigorous analysis is available of both necessary and sufficient conditions for blind PR (for relevant work in 2D see [41], [42], [43] and references therein). In such problems, where an FIR synthesis bank is to be identified, the length (support size) of the synthesis filters is a fundamental issue as it dictates the dimensionality of the unknown parameter space.…”

Using Logarithmic Opinion Pooling Techniques in Bayesian Blind Multi-Channel Restoration

“…See (Sroubek et al, 2007) for a very interesting derivation of the above conditions and for its use in the super resolution problems see . See also (Gastaud et al, 2007) for the possible use of other observation models (regularization terms) for the multi-channel blur.…”