Abstract:The problem of blind image deblurring remains a challenging inverse problem, due to the ill-posed nature of estimating unknown blur kernels and latent images within the Maximum A Posteriori (MAP) framework. To address this challenge, traditional methods often rely on sparse regularization priors to mitigate the uncertainty inherent in the problem. In this paper, we propose a novel blind deblurring model based on the MAP framework that leverages Composite-Gradient Feature (CGF) variations in edge regions after image blurring. This prior term is specifically designed to exploit the high sparsity of sharp edge regions in clear images, thereby effectively alleviating the ill-posedness of the problem. Unlike existing methods that focus on local gradient information, our approach focuses on the aggregation of edge regions, enabling better detection of both sharp and smoothed edges in blurred images. In the blur kernel estimation process, we enhance the accuracy of the kernel by assigning effective edge information from the blurred image to the smoothed intermediate latent image, preserving critical structural details lost during the blurring process. To further improve the edge-preserving restoration, we introduce an adaptive regularizer that outperforms traditional total variation regularization by better maintaining edge integrity in both clear and blurred images. The proposed variational model is efficiently implemented using alternating iterative techniques. Extensive numerical experiments and comparisons with state-of-the-art methods demonstrate the superior performance of our approach, highlighting its effectiveness and real-world applicability in diverse image-restoration tasks.

An Edge-Preserving Blind Image Restoration Using Hierarchical Bayesian Model

Multi-frame Blind Deconvolution Using Sparse Priors

A Retina-Filtering Model and Its Application in Image Restoration

Iterative Blind Image Restoration With Ringing Reduction

A Robust Blind Deblurring Method for Motion Images Based on Edge Information

Blind restoration of compressed degraded image based on block effect suppression

Estimation of blind image restoration using local hierarchical models and MAP

Combined Approach of Blind Image Restoration

A Blind Restoration Method for Remote Sensing Images

Blind Image Restoration Algorithm in Bayesian Framework

High-order Hybrid Regularization Method for Image Blind Restoration

Blind Image Deconvolution Assisted By Robust Non-Gaussian Priors

Choice of Edge-preserving and Sparse Prior Models in Image Restoration

Deep Variational Network Toward Blind Image Restoration

An Edge-Preserving Image Reconstruction Using Neural Network

Blind restoration of real turbulence-degraded image with complicated backgrounds using anisotropic regularization

Semi-blind image restoration based on Canny edge detection

Blind Infrared Remote-Sensing Image Deblurring Algorithm via Edge Composite-Gradient Feature Prior and Detail Maintenance

Sparsity and Smoothing Multi-Regularization Constraints for Blind Image Deblurring

Effective Image Restorations Using a Novel Spatial Adaptive Prior

A BP neural network model for image restoration based on area around pixel and edge information