Wavelet domain based directional dictionaries for single image super-resolution

This paper introduces wavelet domain-based multiple dictionary learning for single image super-resolution. A coupled dictionary learning mechanism is employed using sparsity with the desirable properties of the wavelet transform, such as directionality, analysis at many resolution levels, and persistence across scales. This approach ensures that coupling between signal features (at two consecutive levels) becomes more prominent. We design six pairs of coupled, low and high resolution wavelet sub-band dictionaries. The dictionaries are compact and directional. Super-resolution is achieved by using these low and high resolution dictionaries. Compared to state of the art algorithms, results show significant improvements for the reconstruction of low-resolution images.

[1]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[2]  Michael Elad,et al.  A Plurality of Sparse Representations Is Better Than the Sparsest One Alone , 2009, IEEE Transactions on Information Theory.

[3]  Thomas S. Huang,et al.  Deep Networks for Image Super-Resolution with Sparse Prior , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[4]  Thomas S. Huang,et al.  Coupled Dictionary Training for Image Super-Resolution , 2012, IEEE Transactions on Image Processing.

[5]  Thomas S. Huang,et al.  Image Super-Resolution Via Sparse Representation , 2010, IEEE Transactions on Image Processing.

[6]  Nick G. Kingsbury,et al.  The dual-tree complex wavelet transform: A new efficient tool for image restoration and enhancement , 1998, 9th European Signal Processing Conference (EUSIPCO 1998).

[7]  A. Amanatiadis,et al.  A survey on evaluation methods for image interpolation , 2009 .

[8]  Michael Elad,et al.  Multi-Scale Dictionary Learning Using Wavelets , 2011, IEEE Journal of Selected Topics in Signal Processing.

[9]  Y. C. Pati,et al.  Orthogonal matching pursuit: recursive function approximation with applications to wavelet decomposition , 1993, Proceedings of 27th Asilomar Conference on Signals, Systems and Computers.

[10]  Reinhard Klette,et al.  Sharpness and contrast measures on videos , 2015, 2015 International Conference on Image and Vision Computing New Zealand (IVCNZ).

[11]  Kyoung Mu Lee,et al.  Accurate Image Super-Resolution Using Very Deep Convolutional Networks , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  Quan Pan,et al.  Semi-coupled dictionary learning with applications to image super-resolution and photo-sketch synthesis , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[13]  Stéphane Mallat,et al.  Singularity detection and processing with wavelets , 1992, IEEE Trans. Inf. Theory.

[14]  Richard Baraniuk,et al.  The Dual-tree Complex Wavelet Transform , 2007 .

[15]  Lei Zhang,et al.  Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization , 2010, IEEE Transactions on Image Processing.

[16]  M. Elad,et al.  $rm K$-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation , 2006, IEEE Transactions on Signal Processing.

[17]  Michael Elad,et al.  From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images , 2009, SIAM Rev..

[18]  Mahmoud Nazzal,et al.  Wavelet domain dictionary learning-based single image superresolution , 2015, Signal Image Video Process..

[19]  Madad Ali Shah,et al.  Single image super-resolution by directionally structured coupled dictionary learning , 2016, EURASIP J. Image Video Process..

[20]  G. Bellala.  Characterization of Signals from Multi scale Edges , 2009 .

[21]  A. Amanatiadis,et al.  Performance evaluation techniques for image scaling algorithms , 2008, 2008 IEEE International Workshop on Imaging Systems and Techniques.

[22]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[23]  Reinhard Klette,et al.  Coupled multiple dictionary learning based on edge sharpness for single-image super-resolution , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).