An improved non-blind image deblurring methods based on FoEs
نویسندگان
چکیده
Traditional non-blind image deblurring algorithms always use maximum a posterior(MAP). MAP estimates involving natural image priors can reduce the ripples effectively in contrast to maximum likelihood(ML). However, they have been found lacking in terms of restoration performance. Based on this issue, we utilize MAP with KL penalty to replace traditional MAP. We develop an image reconstruction algorithm that minimizes the KL divergence between the reference distribution and the prior distribution. The approximate KL penalty can restrain over-smooth caused by MAP. We use three groups of images and Harris corner detection to prove our method. The experimental results show that our algorithm of non-blind image restoration can effectively reduce the ringing effect and exhibit the state-of-the-art deblurring results.
منابع مشابه
Blind Image Deblurring via Reweighted Graph Total Variation
Blind image deblurring, i.e., deblurring without knowledge of the blur kernel, is a highly ill-posed problem. The problem can be solved in two parts: i) estimate a blur kernel from the blurry image, and ii) given estimated blur kernel, de-convolve blurry input to restore the target image. In this paper, by interpreting an image patch as a signal on a weighted graph, we first argue that a skelet...
متن کاملBlind deblurring of spiral CT images Based on ENR and Wiener filter
In previous studies, we developed a blind deblurring algorithm based on the edge-to-noise ratio (ENR) and EM algorithm to improve the quality of spiral CT images. In this work, we apply the Wiener filter in place of the EM algorithm to improve the timing performance. The algorithm works well in patient studies. After fully automatic blind deblurring, the conspicuity of the submillimeter feature...
متن کاملLearning a Discriminative Prior for Blind Image Deblurring
We present an effective blind image deblurring method based on a data-driven discriminative prior. Our work is motivated by the fact that a good image prior should favor clear images over blurred images. In this work, we formulate the image prior as a binary classifier which can be achieved by a deep convolutional neural network (CNN). The learned prior is able to distinguish whether an input i...
متن کاملRecent Progress in Image Deblurring
This paper comprehensively reviews the recent development of image deblurring, including nonblind/blind, spatially invariant/variant deblurring techniques. Indeed, these techniques share the same objective of inferring a latent sharp image from one or several corresponding blurry images, while the blind deblurring techniques are also required to derive an accurate blur kernel. Considering the c...
متن کاملGraph-Based Blind Image Deblurring From a Single Photograph
Blind image deblurring, i.e., deblurring without knowledge of the blur kernel, is a highly ill-posed problem. The problem can be solved in two parts: i) estimate a blur kernel from the blurry image, and ii) given estimated blur kernel, deconvolve blurry input to restore the target image. In this paper, we propose a graph-based blind image deblurring algorithm by interpreting an image patch as a...
متن کامل