基于稀疏表示块和全变分的超分辨率重建研究
|
摘要
针对传统的基于稀疏表示超分辨率重建存在不能同时保存边缘与纹理结构,并且在线运行时间长的问题。一种基于稀疏表示块的超分辨重建算法被提出,首先通过图像训练PCA字典集,然后应用PCA算法得到相应的聚类子字典,在重建的过程中引入全变分正则项,以便联系图像局部之间的信息和加强保存重建过程中的图像纹理特征。最后用分步算法求目标函数,重建得到高分辨率图像。大量的仿真实验结果证明,与传统的超分辨率算法相比,新算法能够改善图像结构特征信息,评价指标值也有一定的提高。
For traditional super-resolution reconstruction based on sparse representation it cannot preserve edge and texture well, and those ways have long time for run online. So it is presented algorithm based on sparse representation and total variation for super-resolution reconstruction, first through training PCA dictionary set of images, then applied PCA algorithm to image and get corresponding clustering sub-dictionary. In the process of reconstruction, we added total variation regular item, in order to contact local information in image and preserve texture features of image. Finally, the objective function of high super-resolution reconstruction is got by the segment algorithm. A lot of experiment results show that the new algorithm can both improve the structure feature information of image and increases the evaluation index compared the traditional super-resolution algorithm.
For traditional super-resolution reconstruction based on sparse representation it cannot preserve edge and texture well, and those ways have long time for run online. So it is presented algorithm based on sparse representation and total variation for super-resolution reconstruction, first through training PCA dictionary set of images, then applied PCA algorithm to image and get corresponding clustering sub-dictionary. In the process of reconstruction, we added total variation regular item, in order to contact local information in image and preserve texture features of image. Finally, the objective function of high super-resolution reconstruction is got by the segment algorithm. A lot of experiment results show that the new algorithm can both improve the structure feature information of image and increases the evaluation index compared the traditional super-resolution algorithm.
引文
[1] Yang Jian-chao,Wright J,Huang T S,et al.Image super-resolution via sparse representation[J].IEEE Transactions on Image Processing,2010,19(11):2861-2873.
[2] Zeyde R,Elad M,Protter M.On single image scale-up using sparse-representations[M]//Curves and Surfaces,Springer Berlin Heidelberg,2012:711-730.
[3] Wang S,Zhang D,Liang Y,et al.Semi-coupled dictionary learning with applications to image super-resolution and photo-sketch synthesis[C]//Computer Vision and Pattern Recognition (CVPR),2012:2216-2223.
[4] Takeda H,Farsiu S,Milanfar P.Kernel regression for image processing and reconstruction[J].IEEE Transactions on Image Processing,2007,16(2):349-366.
[5] ZHANG Kaibing,GAO Xinbo,TAO Dacheng,et al.Single image super-resolution with non-local means and steering Kernel regression[J].IEEE Transactions on Image Processing,2012,21(11):4544-4556.
[6] 陈健,高慧斌,王伟国,等.图像超分辨复原方法及应用,激光与光电子学进展.2015,52(2):020004
[7] Xu J,Zhang L,Zuo W,Zhang D,et al.Patch group based nonlocal self-similarity prior learning for image denoising[C],in ICCV 2015.
[8] Huang Y.M,Moisan L,Ng M,Zeng T.Multiplicative noise removal via a learned dictionary[J].IEEE Transactions on Image Processing 2012,21(11):4534-43.
[2] Zeyde R,Elad M,Protter M.On single image scale-up using sparse-representations[M]//Curves and Surfaces,Springer Berlin Heidelberg,2012:711-730.
[3] Wang S,Zhang D,Liang Y,et al.Semi-coupled dictionary learning with applications to image super-resolution and photo-sketch synthesis[C]//Computer Vision and Pattern Recognition (CVPR),2012:2216-2223.
[4] Takeda H,Farsiu S,Milanfar P.Kernel regression for image processing and reconstruction[J].IEEE Transactions on Image Processing,2007,16(2):349-366.
[5] ZHANG Kaibing,GAO Xinbo,TAO Dacheng,et al.Single image super-resolution with non-local means and steering Kernel regression[J].IEEE Transactions on Image Processing,2012,21(11):4544-4556.
[6] 陈健,高慧斌,王伟国,等.图像超分辨复原方法及应用,激光与光电子学进展.2015,52(2):020004
[7] Xu J,Zhang L,Zuo W,Zhang D,et al.Patch group based nonlocal self-similarity prior learning for image denoising[C],in ICCV 2015.
[8] Huang Y.M,Moisan L,Ng M,Zeng T.Multiplicative noise removal via a learned dictionary[J].IEEE Transactions on Image Processing 2012,21(11):4534-43.