利用低秩先验的噪声模糊图像盲去卷积
Blind Deconvolution for Noisy and Blurry Images Using Low Rank Prior
查看参考文献22篇
文摘
|
单幅图像盲去卷积的目的是从一幅观测的模糊图像估计出模糊核和清晰图像。该问题是严重病态的,尤其是观测图像中噪声不可忽略时更具挑战性。该文主要针对如何有效利用低秩先验约束进行噪声模糊图像盲去卷积问题,提出一种在交替最大后验(MAP)估计框架下利用低秩先验约束的单幅噪声模糊图像盲去卷积方法。首先,在估计中间复原图像时,利用低秩先验约束对复原图像中的噪声进行抑制。然后,采用降噪后的中间复原图像估计模糊核,得到更好质量的模糊核估计。迭代上述两个操作获得最终可靠的模糊核估计。最后,根据所估计的模糊核,通过非盲去卷积方法复原出清晰图像。实验结果表明:所提方法在定量和定性评价指标上优于已有的代表性方法。 |
其他语种文摘
|
The purpose of single image blind deconvolution is to estimate the unknown blur kernel from a single observed blurred image and recover the original sharp image. Such a task is severely ill-posed and even more challenging especially in the condition that the noise in the input image can not be negligible. In this paper, the main problem this study focuses on is how to effectively apply low rank prior to blind deconvolution. A single noisy and blurry image blind deconvolution algorithm is proposed, using alternating Maximum A Posteriori (MAP) estimation combined with low rank prior. First, when estimating the intermediate latent image, low rank prior is used as the constraint that is used for noise suppression of the restored image. Then the denoised intermediate latent image in turn leads to higher quality blur kernel estimation. These two operations are iterated in this manner to arrive at reliable blur kernel estimation. Finally, the non-blind deconvolution method is chosen to be used with sparse prior knowledge to achieve the final latent image restoration. Extensive experiments manifest the superiority of the proposed method over state-of-the-art techniques, both qualitatively and quantitatively. |
来源
|
电子与信息学报
,2017,39(8):1919-1926 【核心库】
|
DOI
|
10.11999/jeit161206
|
关键词
|
盲去卷积
;
最大后验估计
;
噪声模糊图像
;
低秩先验
|
地址
|
中国科学院沈阳自动化研究所光电信息技术研究室, 中国科学院光电信息处理重点实验室;;辽宁省图像理解与视觉计算重点实验室, 沈阳, 110016
|
语种
|
中文 |
文献类型
|
研究性论文 |
ISSN
|
1009-5896 |
学科
|
自动化技术、计算机技术 |
基金
|
辽宁省教育厅科学研究计划项目
|
文献收藏号
|
CSCD:6045149
|
参考文献 共
22
共2页
|
1.
Levin A. Efficient marginal likelihood optimization in blind deconvolution.
IEEE Conference on Computer Vision and Pattern Recognition,2011:2657-2664
|
被引
7
次
|
|
|
|
2.
Krishnan D. Blind deconvolution using a normalized sparsity measure.
IEEE Conference on Computer Vision and Pattern Recognition,2011:233-240
|
被引
12
次
|
|
|
|
3.
Sun L. Edge-based blur kernel estimation using patch priors.
IEEE Conference on Computational Photography,2013:1-8,19-21
|
被引
2
次
|
|
|
|
4.
Lai W S. Blur kernel estimation using normalized color-line priors.
IEEE Conference on Computer Vision and Pattern Recognition,2015:64-72
|
被引
2
次
|
|
|
|
5.
Ren W. Image deblurring via enhanced low-rank prior.
IEEE Transactions on Image Processing,2016,25(7):3426-3437
|
被引
4
次
|
|
|
|
6.
Cho S. Fast motion deblurring.
ACM Transactions on Graphics,2009,28(5):89-97
|
被引
91
次
|
|
|
|
7.
Xu L. Two-phase kernel estimation for robust motion deblurring.
European Conference on Computer Vision,2010:157-170
|
被引
13
次
|
|
|
|
8.
Pan J. Kernel estimation from salient structure for robust motion deblurring.
Signal Processing: Image Commmunication,2013,28(9):1156-1170
|
被引
8
次
|
|
|
|
9.
Pan J. L0-regularized intensity and gradient prior for text images deblurring and beyond.
IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,39(2):342-355
|
被引
2
次
|
|
|
|
10.
Pan J. Blind image deblurring using dark channel prior.
IEEE Conference on Computer Vision and Pattern Recognition,2016:1628-1636
|
被引
5
次
|
|
|
|
11.
Mai L. Kernel fusion for better image deblurring.
IEEE Conference on Computer Vision and Pattern Recognition,2015:371-380
|
被引
1
次
|
|
|
|
12.
Yan R M. Blind image blur estimation via deep learning.
IEEE Transactions on Image Processing,2016,25(4):1910-1921
|
被引
11
次
|
|
|
|
13.
Tai Y. Motion-aware noise filtering for deblurring of noisy and blurry images.
IEEE Conference on Computer Vision and Pattern Recognition,2012:17-24
|
被引
1
次
|
|
|
|
14.
Zhong L. Handling noise in single image deblurring using directional filters.
IEEE Conference on Computer Vision and Pattern Recognition,2013:612-619
|
被引
2
次
|
|
|
|
15.
Whyte O. Non-uniform deblurring for shaken images.
IEEE Conference on Computer Vision and Pattern Recognition,2010:491-498
|
被引
1
次
|
|
|
|
16.
Gu S. Weighted nuclear norm minimization with application to image denoising.
IEEE Conference on Computer Vision and Pattern Recognition,2014:2862-2869
|
被引
6
次
|
|
|
|
17.
Dong W S. Nonlocal image restoration with bilateral variance estimation: A low-rank approach.
IEEE Transactions on Image Processing,2013,22(2):700-711
|
被引
36
次
|
|
|
|
18.
Cai J F. A singular value thresholding algorithm for matrix completion.
SIAM Journal on Optimization,2008,20(4):1956-1982
|
被引
34
次
|
|
|
|
19.
Nocedal J.
Numerical Optimization,2006:497-506
|
被引
2
次
|
|
|
|
20.
Levin A. Understanding and evaluating blind deconvolution algorithms.
IEEE Conference on Computer Vision and Pattern Recognition,2009:1964-1971
|
被引
14
次
|
|
|
|
|