|
|
- 2016
联合新型分块稀疏表示和梯度先验图像盲复原
|
Abstract:
针对目前基于稀疏表示的图像盲复原算法计算量大且细节恢复能力有限等问题, 提出一种新的图像盲复原方法．首先针对现有稀疏表示模型中重叠分块计算复杂度高的问题, 提出一种多模式非重叠分块策略, 在每种模式下独立求解复原图像, 然后对各模式下复原图像求平均以消除“伪像”; 另外, 用范数作为稀疏性度量, 将图像梯度稀疏先验融入基于稀疏表示的图像盲复原模型．最后, 本文提出了联合新型分块字典稀疏表示和图像梯度稀疏先验的盲复原模型, 采取迭代方法交替估计模糊核和待复原图像．实验结果表明, 该方法在主观和客观评价下均取得较好的复原结果, 并显著降低算法整体复杂度．
In the light of the heavy computational loads and the limited ability in detail preserving of sparse representation-based blind image deblurring algorithms，a new method based on patch-dictionary approach and gradient prior was proposed in this paper. First，in order to increase the computational efficiency，a multimode non-overlapping strategy，in which the image recovery was independently carried out in each mode，was presented，and then the results of the subproblems were averaged to eliminate artificial effects; In addition，using the norm as the sparsity measure，we incorporated the gradient prior into the sparse deblurring model. Finally，we designed a novel combined new patch-dictionary sparse representation and gradient prior image deblurring model in which the deblur kernel updating and the deblurred image estimating were performed in turn via the iteration. Experimental results show that the approach proposed achieves better results in both subjective and objective evaluation criteria and significantly reduces the overall complexity of the algorithm
| [1] | Krishnan D, Fergus R. Fast image deconvolution using hyper-Laplacian priors [C]//<i>Advances in Neural Information Processing System</i>. Vancouver, Canada, 2009:1033-1041. |
| [2] | Yang Aiping, Zhong Tengfei, He Yuqing. Super-resolution reconstruction based on non-local similarity and clustered semi-coupled dictionary learning [J]. <i>Journal of Tianjin University</i>:<i>Science and Technol-ogy</i>, 2015, 48(1):87-94(in Chinese). |
| [3] | 程广涛, 宋占杰, 陈雪. 基于二维图像矩阵的稀疏表示分类方法[J]. 天津大学学报:自然科学与工程技术版, 2014, 47(6):541-545. |
| [4] | Cheng Guangtao, Song Zhanjie, Chen Xue. Classification based on sparse representation with two dimensional image matrix [J]. <i>Journal of Tianjin University</i>:<i>Science and Technology</i>, 2014, 47(6):541-545(in Chinese). |
| [5] | Hu Z, Huang J B, Yang M H. Single image deblurring with adaptive dictionary learning[C]// <i>IEEE International</i> <i>Conference on Image Processing</i>. Piscataway, NJ, United States, 2010:1169-1172. |
| [6] | Zhang H, Yang J, Zhang Y, et al. Sparse representation based blind image deblurring [C]// <i>IEEE International Conference on Multimedia and Expo</i>. Barcelona, Spain, 2011:1-6. |
| [7] | Jia C, Evans B L. Patch-based image deconvolution via joint modeling of sparse priors [C]// <i>IEEE International Conference on Information Processing</i>. Brussels, Belgium, 2011:681-684. |
| [8] | Li H, Zhang Y, Zhang H, et al. Blind image deblurring based on sparse prior of dictionary pair [C]// <i>IEEE International Conference on Pattern Recognition</i>. Tsukuba, Japan, 2012:3054-3057. |
| [9] | Roth S, Black M J. Fields of experts:A framework for learning image priors [C]// <i>IEEE International Conference on Computer Vision and Pattern Recognition</i>. Boston, USA, 2005:860-867. |
| [10] | Taeg S C, Joshi N, Zitnick Z L, et al. A content-aware image prior [C]// <i>IEEE International Conference on Computer Vision and Pattern Recognition</i>. San Francisco, USA, 2012:3054-3057. |
| [11] | Shan Q, Jia J Y, Agarwala A. High-quality motion deblurring from a single image [J]. <i>ACM Transactions on Graphics</i>, 2008, 27(3):1-10. |
| [12] | Elad M, Aharon M. Image denoising via sparse and redundant representations over learned dictionaries[J]. <i>IEEE Transactions on Image Processing</i>, 2006, 15(12):3736-3745. |
| [13] | Xu Yangyang, Yin Wotao. A Fast Patch-Dictionary Method for Whole Image Recovery. CAM-Report-13-38 [R]. Los Angeles, CA:UCLA, 2013. |
| [14] | Zhang Y, Yang J, Wotao Y. YALL1:Your Algorithms for l1, MATLAB Software[EB/OL]. http://yall1. blogs. rice. edu/, 2010-07-15. |
| [15] | Levin A, Weiss Y, Durand F, et al. Understanding and evaluating blind deconvolution algorithms [C]//<i>IEEE International Conference on Computer Vision and Pattern Recognition</i>. Miami, USA, 2009:1964-1971. |
| [16] | Levin A, Weiss Y, Durand F, et al. Efficient marginal likelihood optimization in blind deconvolution [C]// <i>IEEE Conference on Computer Vision and Pattern Recognition</i>. Providence, USA, 2011:2657-2664. |
| [17] | Babacan S D, Molina R, Do M N, et al. Bayesian blind deconvolution with general sparse image priors [C]// <i>Computer Vision</i>-<i>ECCV</i> 2012. Berlin, Germany, 2012, 7577:341-355. |
| [18] | Wiener N. <i>Extrapolation</i>, <i>Interpolation</i>, <i>and Smoothing of Stationary Time Series</i> [M]. Cambridge:MIT Press, 1964. |
| [19] | Aharon M, Elad M, Bruckstein A. K-SVD:An algorithm for designing overcomplete dictionaries for sparse representation [J]. <i>IEEE Transactions on Signal Processing</i>, 2006, 54(11):4311-4322. |
| [20] | Lucy L B. An iterative technique for the rectification of observed distributions [J]. <i>Astronomical Journal</i>, 1974, 79(6):745-754. |
| [21] | Fergus R, Singh B, Hertzmann P. Removing camera shake from a single photograph [J]. <i>ACM Transactions on Graphics</i>, 2006, 25(3):787-794. |
| [22] | Krishnan D, Tay T, Fergus R. Blind deconvolution using a normalized sparsity measure [C]//<i>IEEE International Conference on Computer Vision and Pattern Recognition</i>. Colorado, USA, 2011:233-240. |
| [23] | 杨爱萍, 钟腾飞, 何宇清. 基于非局部相似性和分类半耦合字典学习的超分辨率重建[J]. 天津大学学报:自然科学与工程技术版, 2015, 48(1):87-94. |
| [24] | Mairal J, Bach F, Ponce J, et al. Online dictionary learning for sparse coding [C]// <i>International Conference on Machine Learning</i>. Montreal, Canada:ACM, 2009:689-696. |
