Abstract
Blind image deblurring, composed of estimating blur kernel and non-blind deconvolution, is an extremely ill-posed problem. However, previous deblurring methods still cannot solve delta kernel or noise problem well and avoid ringing artifacts in restored image, because the reliable kernel estimation and image restoration could not be given from information-deficiency input without using natural image priors. In this work, we find that the blur process changes the distribution of the image gradient, and thus attempt to use the priori knowledge for guiding blind deblurring. For the sake of convenience of modeling, we come up with a simplified approximate formulation of the image gradient distribution prior, thus the restoration model using it can be solved by the method of iteratively reweighted least squares (IRLS). We also concentrate on how to optimize the models and develop an algorithm based on gradient prior and image structure: first, we computed image structure based on the total variation model and adaptive weight strategy and then estimated the strong edges from it. Those strong edges (structure) that have a possible adverse effect on blur kernel estimation can be removed. Next, we followed an alternate iterative framework to obtain high-quality blur kernel estimation by estimating blur kernel from strong structure and then restoring a latent image guided by the formulated clear image priors. Finally, we proposed a non-blind deconvolution method based on fitted multi-order gradient priors as regularization to restore the latent image. In addition, we analyze the effectiveness of the knowledge-driven prior in image deblurring and demonstrate that it can favor clear images over blurred ones and restrain ringing artifacts effectively. Extensive experiments verify the superiority of the proposed method over state-of-the-art algorithms compared, both qualitatively and quantitatively.
Highlights
Image quality is critical in the applications of intelligent security, biomedical imaging, astronomical observations, and remote sensing
Considering that the delta kernel and noise are introduced by improper image modeling and regularization, we propose a knowledge-driven gradient prior, which learns from limited training images and is a powerful tool for modeling the signal, and we define it as knowledge-driven prior (KDP) in this paper
In this paper, we introduce an image gradient distribution priors-inspired restoration algorithm, which proves to be effective for blind image deblurring by accurately estimating blur kernel and removing ringing artifacts
Summary
Image quality is critical in the applications of intelligent security, biomedical imaging, astronomical observations, and remote sensing. We can get the high-quality image from the estimated blur kernel k∗, noise n∗ and assumed blurry image B∗, by using existing non-blind deconvolution methods (e.g. wiener filtering [30]). For the sake of recovering potential sharp image while estimating the blur kernel, general solutions usually develop regularization of prior for modifying restoration model. We introduce an algorithm for solving latent image restoration model with KDP, which suppresses the kernel noise, ringing effects and elevates the performance of deblurring approaches. For removing blur kernel noises and obtaining accurate kernel estimation, we propose an effective discriminative prior, which is modeled by the clear image gradient distribution. During non-blind deconvolution, to recover high-quality visual effects and suppress ringing artifacts, we propose to use clear image gradient and second-order gradient distribution knowledge as image priors.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
