|
|
- 2017
基于盲反卷积的超分辨率图像盲复原算法
|
Abstract:
为解决超分辨率图像盲复原问题, 研究了一种广义图像降质模型及广义超分辨率图像盲复原算法模型. 提出了基于TVBD的交替最小化超分辨率图像盲复原算法, 并加以改进以改善复原效果; 通过结合MSAA算法, 提出了基于TV交替最小化的快速中值超分辨率图像盲复原算法, 排除野值干扰, 提高算法精度和速度; 根据峰值信噪比和误差平方和, 提出两种新的客观评价指标, 衡量各算法的复原效果. 实验表明, 本文算法有效实现了超分辨率图像盲复原, 并提高了复原精度.
In order to solve the problem in blind super resolution image reconstruction(BSRIR)，a novel general image degradation model and a general BSRIR model were studied. A BSRIR algorithm based on total variation blind deconvolution(TVBD)and alternating minimization(AM)was proposed and modified to improve the reconstruction. By combining MSAA with the algorithm mentioned above，a fast BSRIR algorithm based on TVBD and MSAA was proposed to remove the outlier effect and improve the speed and accuracy. According to peak signal to noise ratio(PSNR)and sum of square difference(SSD)，two novel objective quality metrics were proposed to assess the BSRIR algorithms. Experimental results confirm that the proposed algorithm can achieve higher accuracy in BSRIR
| [1] | 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 Technology</i>, 2015, 48(1):87-94(in Chinese). |
| [2] | 孙云山, 张立毅, 段继忠. 医学CT图像恒模盲均衡算法性能分析[J]. 天津大学学报, 2011, 44(12):1058-1062. |
| [3] | Sun Yunshan, Zhang Liyi, Duan Jizhong. Performance |
| [4] | analysis of constant module nedical CT image blind equalization algorithm[J]. <i>Journal of Tianjin Univer-sity</i>, 2011, 44(12):1058-1062 (in Chinese). |
| [5] | 段继忠, 张立毅, 刘昱, 等. 基于自一致性的磁共振并行成像高效重构算法[J]. 天津大学学报:自然科学与工程技术版, 2014, 47(5):414-419. |
| [6] | Duan Jizhong, Zhang Liyi, Liu Yu, et al. Efficient reconstruction algorithm for parallel magnetic resonance imaging based on self-consistency[J]. <i>Journal of Tianjin University</i>:<i>Science and Technology</i>, 2014, 47(5):414-419(in Chinese). |
| [7] | Marquina A, Osher S J. Image super-resolution by TV-regularization and bregman iteration [J]. <i>Journal of Scientific Computing</i>, 2008, 37(3):367-382. |
| [8] | Irani M, Peleg S. Improving resolution by image registration [J]. <i>Graphical Models and Image Processing</i>, 1991, 53(3):231-239. |
| [9] | 卓力, 王素玉, 李晓光. 图像/视频的超分辨率复原[M]. 北京:人民邮电出版社, 2011. |
| [10] | Freeman W T, Jones T R, Pasztor E C. Example-based super-resolution [J]. <i>IEEE Computer Graphics and Applications, </i>2002, 22(2):56-65. |
| [11] | Levin A, Weiss Y, Durand F, et al. Understanding and evaluating blind deconvolution algorithms[C]// <i>Proceedings of IEEE Conference on Computer Vision and Pattern Recognition</i>. Miami, USA, 2009:1964-1971. |
| [12] | Eldar Y C. Generalized SURE for exponential families:Applications to regularization [J]. <i>IEEE Transactions on Signal Processing, </i>2009, 57(2):471-481. |
| [13] | Ramani S, Liu Z, Rosen J, et al. Regularization parameter selection for nonlinear iterative image restoration and MRI reconstruction using GCV and SURE-based methods [J]. <i>IEEE Transactions on Image Processing</i>, 2012, 21(8):3659-3672. |
| [14] | Elad M, Feuer A. Restoration of a single superresolution image from several blurred, noisy, and undersampled measured images [J]. <i>IEEE Transactions on Image Processing</i>, 1997, 6(12):1646-1658. |
| [15] | Rajan D, Chaudhuri S, Joshi M V. Multi-objective super resolution:Concepts and examples [J]. <i>IEEE Signal Processing Magazine</i>, 2003, 20(3):49-61. |
| [16] | Zhuo Li, Wang Suyu, Li Xiaoguang. <i>Image</i>/<i>Video Super Resolution</i>[M]. Beijing:Posts & Telecom Press, 2011(in Chinese). |
| [17] | 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. |
| [18] | Sun Yunshan, Zhang Liyi, Zhang Haiyan, et al. Multichannel blind CT image restoration via variable splitting and alternating direction method[J]. <i>Transactions of Tianjin University</i>, 2015, 21(6):524-532. |
| [19] | Ichigaya A, Kurozumi M, Hara N, et al. A method of estimating coding PSNR using quantized DCT coefficients [J]. <i>IEEE Transactions on Circuits and Systems for Video Technology</i>, 2006, 16(2):251-259. |
| [20] | Thompson A M, Brown J C, Kay J W, et al. A study of methods of choosing the smoothing parameter in image restoration by regularization [J]. <i>IEEE Transactions on Pattern Analysis & Machine Intelligence</i>, 1991(4):326-339. |
| [21] | Farsiu S, Robinson M D, Elad M, et al. Fast and robust multiframe super resolution [J]. <i>IEEE Transactions on Image Processing</i>, 2004, 13(10):1327-1344. |
| [22] | Chan T F, Wong C K. Total variation blind deconvolution [J]. <i>IEEE Transactions on Image Processing</i>, 1998, 7(3):370-375. |
| [23] | Perrone D, Favaro P. Total variation blind deconvolu-tion:The devil is in the details [C]// <i>Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition</i>. Columbus, USA, 2014:2909-2916. |
| [24] | Levin A, Fergus R, Durand F, et al. Image and depth from a conventional camera with a coded aperture[J]. <i>ACM Transactions on Graphics</i>, 2007, 26(3):70. |
| [25] | Harmeling S, Sra S, Hirsch M, et al. Multiframe blind deconvolution, super-resolution, and saturation correction via incremental EM [C]// <i>Proceedings of the<i> 17 <i>th IEEE International Conference on Image Processing</i>. Hongkong, China, 2010:3313-3316.</i></i> |
