基于多字典的单幅图像超分辨率重建
|
摘要
对于目前图像超分辨率重建算法中的问题,忽略重建图像结构性和重建过程中丢失高频信息,提出了一种基于多字典的单幅图像超分辨率重建算法。在字典学习阶段根据每个图像块的主方向角,对所有训练图像块进行聚类并训练各类的字典。利用训练得到的字典重建训练样本并计算各类的残差图像块,然后对残差图像块再进行聚类、训练残差字典。用锚定邻域回归方法重建高分辨率图像,实验结果表明,该算法在客观评价和视觉效果上均优于许多优秀的图像超分辨算法。
This paper presents a new multi-dictionary learning algorithm for the problem of single super-resolution for the common dictionary-based SR algorithm ignoring the structure of the image, and some high-frequency information in the image reconstruction process. Firstly, the dominant direction of each image patch should be computed before clustering, then training dictionaries. Next, the residual image patches which computed by the dictionaries can be used to get residual dictionaries. Last, adjusted anchored neighborhood regression is used to reconstruct image, experimental results show that the algorithm has a better result both on visual effect and objective compared with other methods.
This paper presents a new multi-dictionary learning algorithm for the problem of single super-resolution for the common dictionary-based SR algorithm ignoring the structure of the image, and some high-frequency information in the image reconstruction process. Firstly, the dominant direction of each image patch should be computed before clustering, then training dictionaries. Next, the residual image patches which computed by the dictionaries can be used to get residual dictionaries. Last, adjusted anchored neighborhood regression is used to reconstruct image, experimental results show that the algorithm has a better result both on visual effect and objective compared with other methods.
引文
[1]Freeman T,Jones T R,Pastor E C.Example based superresolution[J].IEEE Computer Graphics and Applications,2002,22(2):56-65.
[2]Yang J,Wright J,Huang T,et al.Image super-resolution as sparse representation of raw image patches[C]//CVPR,2008.
[3]Yang J,Wright J,Huang T,et al.Image super-resolution via sparse representation[J].IEEE Trans on Image Processing,2010,19(11):2861-2873.
[4]Zeyde R,Elad E,Protter M.On single image scale-up using sparse-representations,curves and surfaces[M].Berlin,Germany:Springer,2012:711-730.
[5]Timofte R,Smet V D,Gool L V.Anchored neighborhood regression for fast example-based super resolution[C]//ICCV,2013:1920-1927.
[6]Timofte R,Smet V D,Gool L V.A+:adjusted anchored neighborhood regression for fast super-resolution[C]//Asian Conference on Computer Vision.Cham:Springer,2014:111-126.
[7]Ahmed J,Shah M A.Single image super-resolution by directionally structured coupled dictionary learning[J].EURASIP Journal on Image&Video Processing,2016.
[8]Farhadifard F,Abar E,Nazzal M,et al.Single image super resolution based on sparse representation via directionally structured dictionaries[C]//Signal Processing and Communications Applications Conference,2014:1718-1721.
[9]Yeganli F,Nazzal M,Ozkaramanli H.Super-resolution using multiple structured dictionaries based on the gradient operator and bicubic interpolation[C]//Signal Processing and Communication Application Conference,2016:941-944.
[10]Yang S,Wang M,Chen Y,et al.Single-image superresolution reconstruction via learned geometric dictionaries and clustered sparse coding[J].IEEE Transactions on Image Processing,2012,21(9):4016-4028.
[11]Zhang Y,Wu W,Dai Y,et al.Remote sensing images super-resolution based on sparse dictionaries and residual dictionaries[C]//IEEE International Conference on Dependable,Autonomic and Secure Computing,2013:318-323.
[12]Nazzal M,Yeganli F,Ozkaramanli H.A strategy for residual component-based multiple structured dictionary learning[J].IEEE Signal Processing Letters,2015,22(11):2059-2063.
[13]Nazzal M,Yeganli F,Ozkaramanli H.Improved dictionary learning by constrained re-training over residual components[C]//Signal Processing and Communication Application Conference,2016:833-836.
[14]Dong C,Loy C C,He K,et al.Learning a deep convolutional network for image super-resolution[C]//ECCV,2014:184-199.
[15]Song S,Li Y,Liu J,et al.Joint sub-band based neighbor embedding for image super-resolution[C]//IEEEInternational Conference on Acoustics,Speech and Signal Processing,2016:1661-1665.
[16]Huang J B,Singh A,Ahuja N.Single image super-resolution from transformed self-exemplars[C]//Proc IEEEInt’l Conf Computer Vision and Pattern Recognition,2015:5197-5206.
[17]Zhang K,Li J,Wang H,et al.Learning local dictionaries and similarity structures for single image super-resolution[J].Signal Processing,2017.
[18]Dai D,Timofte R,Van Gool L.Jointly optimized regressors for image super-resolution[C]//EUROGRAPHICS,2015:95-104.
[19]Zhang K,Li J,Xiong Z,et al.Optimized multiple linear mappings for single image super-resolution[J].Optics Communications,2017.
[2]Yang J,Wright J,Huang T,et al.Image super-resolution as sparse representation of raw image patches[C]//CVPR,2008.
[3]Yang J,Wright J,Huang T,et al.Image super-resolution via sparse representation[J].IEEE Trans on Image Processing,2010,19(11):2861-2873.
[4]Zeyde R,Elad E,Protter M.On single image scale-up using sparse-representations,curves and surfaces[M].Berlin,Germany:Springer,2012:711-730.
[5]Timofte R,Smet V D,Gool L V.Anchored neighborhood regression for fast example-based super resolution[C]//ICCV,2013:1920-1927.
[6]Timofte R,Smet V D,Gool L V.A+:adjusted anchored neighborhood regression for fast super-resolution[C]//Asian Conference on Computer Vision.Cham:Springer,2014:111-126.
[7]Ahmed J,Shah M A.Single image super-resolution by directionally structured coupled dictionary learning[J].EURASIP Journal on Image&Video Processing,2016.
[8]Farhadifard F,Abar E,Nazzal M,et al.Single image super resolution based on sparse representation via directionally structured dictionaries[C]//Signal Processing and Communications Applications Conference,2014:1718-1721.
[9]Yeganli F,Nazzal M,Ozkaramanli H.Super-resolution using multiple structured dictionaries based on the gradient operator and bicubic interpolation[C]//Signal Processing and Communication Application Conference,2016:941-944.
[10]Yang S,Wang M,Chen Y,et al.Single-image superresolution reconstruction via learned geometric dictionaries and clustered sparse coding[J].IEEE Transactions on Image Processing,2012,21(9):4016-4028.
[11]Zhang Y,Wu W,Dai Y,et al.Remote sensing images super-resolution based on sparse dictionaries and residual dictionaries[C]//IEEE International Conference on Dependable,Autonomic and Secure Computing,2013:318-323.
[12]Nazzal M,Yeganli F,Ozkaramanli H.A strategy for residual component-based multiple structured dictionary learning[J].IEEE Signal Processing Letters,2015,22(11):2059-2063.
[13]Nazzal M,Yeganli F,Ozkaramanli H.Improved dictionary learning by constrained re-training over residual components[C]//Signal Processing and Communication Application Conference,2016:833-836.
[14]Dong C,Loy C C,He K,et al.Learning a deep convolutional network for image super-resolution[C]//ECCV,2014:184-199.
[15]Song S,Li Y,Liu J,et al.Joint sub-band based neighbor embedding for image super-resolution[C]//IEEEInternational Conference on Acoustics,Speech and Signal Processing,2016:1661-1665.
[16]Huang J B,Singh A,Ahuja N.Single image super-resolution from transformed self-exemplars[C]//Proc IEEEInt’l Conf Computer Vision and Pattern Recognition,2015:5197-5206.
[17]Zhang K,Li J,Wang H,et al.Learning local dictionaries and similarity structures for single image super-resolution[J].Signal Processing,2017.
[18]Dai D,Timofte R,Van Gool L.Jointly optimized regressors for image super-resolution[C]//EUROGRAPHICS,2015:95-104.
[19]Zhang K,Li J,Xiong Z,et al.Optimized multiple linear mappings for single image super-resolution[J].Optics Communications,2017.