Prior-guided deep residual propagation for non-blind image deconvolution
Authors: 
Long Ma, Zhiying Jiang, Risheng Liu, Xin FanAuthors Info & Claims
Long Ma, Zhiying Jiang, Risheng Liu, Xin FanAuthors Info & ClaimsArticle No.: 38, Pages 1 - 5
Abstract
Non-blind image deconvolution aims to recover clear images from observations corrupted by blur kernel, which plays a very important role in many signal processing and computer vision applications. In recent years, deep convolutional neural networks (CNNs) have achieved good performance on low-level image processing areas, such as denoising and superresolution. But unfortunately, till now only a few CNNs have been proposed for image deconvolution. It is mainly because this task is highly ill-posed and often with less high-quality training data. In this paper, we develop a novel prior-guided residual CNN to integrate task formulation, our cues on the incorporation of domain knowledge and data-dependent network for non-blind image deconvolution. The image propagation generated by our framework indeed gains advantages from both physical principles of the task and training data, thus can successfully recover more accurate deconvolution results. Experimental results on various challenging benchmarks demonstrate that our method can obtain excellent performance quantitatively and qualitatively.
References
[1]
Hédy Attouch, Jérôme Bolte, Patrick Redont, and Antoine Soubeyran. 2010. Proximal alternating minimization and projection methods for nonconvex problems: an approach based on the Kurdyka-Łojasiewicz inequality. Mathematics of Operations Research 35, 2 (2010), 438--457.
[2]
Yunjin Chen and Thomas Pock. 2017. Trainable nonlinear reaction diffusion: A flexible framework for fast and effective image restoration. IEEE Transaction on Pattern Analysis and Machine Intelligence 39, 6 (2017), 1256--1272.
[3]
Aram Danielyan, Vladimir Katkovnik, and Karen Egiazarian. 2012. BM3D frames and variational image deblurring. IEEE Transaction on Image Processing 21, 4 (2012), 1715--1728.
[4]
Rob Fergus, Barun Singh, Aaron Hertzmann, Sam T. Roweis, and William T. Freeman. 2006. Removing camera shake from a single photograph. Acm Transactions on Graphics 25, 3 (2006), 787--794.
[5]
Dilip Krishnan and Rob Fergus. 2009. Fast image deconvolution using hyper-Laplacian priors. In Advances in Neural Information Processing Systems. 1033--1041.
[6]
Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. 2012. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems. 1097--1105.
[7]
Stamatios Lefkimmiatis. 2017. Non-local color image denoising with convolutional neural networks. In Proceeding of IEEE Conference on Computer Vision and Pattern Recognition.
[8]
Anat Levin, Yair Weiss, Fredo Durand, and William T. Freeman. 2009. Understanding and evaluating blind deconvolution algorithms. In Proceeding of IEEE Conference on Computer Vision and Pattern Recognition Recognition. 1964--1971.
[9]
Tim Meinhardt, Michael Möller, Caner Hazirbas, and Daniel Cremers. 2017. Learning Proximal Operators: Using Denoising Networks for Regularizing Inverse Imaging Problems. In Proceeding of IEEE Conference on Computer Vision and Pattern Recognition.
[10]
Pietro Perona and Jitendra Malik. 1990. Scale-space and edge detection using anisotropic diffusion. IEEE Transactions on pattern analysis and machine intelligence 12, 7 (1990), 629--639.
[11]
Stefan Roth and Michael J. Black. 2009. Fields of Experts. International Journal of Computer Vision 82, 2 (2009), 205--229.
[12]
Uwe Schmidt and Stefan Roth. 2014. Shrinkage Fields for Effective Image Restoration. In Proceeding of IEEE Conference on Computer Vision and Pattern Recognition. 2774--2781.
[13]
Christian J Schuler, Harold Christopher Burger, Stefan Harmeling, and Bernhard Scholkopf. 2013. A Machine Learning Approach for Non-blind Image Deconvolution. In Proceeding of IEEE Conference on Computer Vision and Pattern Recognition. 1067--1074.
[14]
Libin Sun, Sunghyun Cho, Jue Wang, and James Hays. 2013. Edge-based blur kernel estimation using patch priors. In IEEE International Conference on Computational Photography. 1--8.
[15]
Joachim Weickert. 1998. Anisotropic diffusion in image processing. Teubner Stuttgart.
[16]
L. Xu, J. SJ. Ren, C. Liu, and J. Jia. 2014. Deep convolutional neural network for image deconvolution. In Advances in Neural Information Processing Systems. 1790--1798.
[17]
Kai Zhang, Wangmeng Zuo, Yunjin Chen, Deyu Meng, and Lei Zhang. 2016. Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising. IEEE Transaction on Image Processing 26, 7 (2016), 3142--3155.
[18]
Kai Zhang, Wangmeng Zuo, Shuhang Gu, and Lei Zhang. 2017. Learning Deep CNN Denoiser Prior for Image Restoration. In Proceeding of IEEE Conference on Computer Vision and Pattern Recognition.
[19]
Daniel Zoran and Yair Weiss. 2011. From learning models of natural image patches to whole image restoration. In Proceeding of IEEE Conference on Computer Vision and Pattern Recognition. 479--486.
Index Terms
- Prior-guided deep residual propagation for non-blind image deconvolution
Recommendations
High-quality non-blind image deconvolution with adaptive regularization
Non-blind image deconvolution is a process that obtains a sharp latent image from a blurred image when a point spread function (PSF) is known. However, ringing and noise amplification are inevitable artifacts in image deconvolution since perfect PSF ...
Non-blind image deconvolution using a regularization based on re-blurring process
Highlights- We propose a non-blind image deconvolution method, with a new image prior that favors sharp images over blurry ones.
AbstractImage deconvolution is an ill-posed problem that requires a regularization term to solve. The most common forms of image priors used as the regularization term in image deconvolution tend to produce smoothed (slightly blurry) images, ...
Non-blind image deconvolution with adaptive regularization
PCM'10: Proceedings of the 11th Pacific Rim conference on Advances in multimedia information processing: Part IRinging and noise amplification are the most dominant artifacts in image deconvolution. These artifacts can be reduced by introducing image prior into the deconvolution process. A regularization weighting factor can control the strength of ...
Comments
Information & Contributors
Information
Published In
August 2018
243 pages
ISBN:9781450365208
DOI:10.1145/3240876
Copyright © 2018 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 17 August 2018
Check for updates
Author Tags
Qualifiers
- Research-article
Funding Sources
- National Natural Science Foundation of China
- Fundamental Research Funds for the Central Universities
Conference
ICIMCS'18
ICIMCS'18: The 10th International Conference on Internet Multimedia Computing and Service
August 17 - 19, 2018
Nanjing, China
Acceptance Rates
ICIMCS '18 Paper Acceptance Rate 46 of 116 submissions, 40%;
Overall Acceptance Rate 163 of 456 submissions, 36%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 64Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 31 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in