Sparse representation using multiple dictionaries for single image super-resolution

New algorithms are proposed in this paper for single image super-resolution using multiple dictionaries based on sparse representation. In the proposed algorithms, a classifier is constructed which is based on the edge properties of image patches via the two lowest discrete cosine transformation (DCT) coefficients. The classifier partitions all training patches into three classes. Training patches from each of the three classes can then be used for the training of the corresponding dictionary via the K-SVD (singular value decomposition) algorithm. Experimental results show that the high resolution image quality using the proposed algorithms is better than that using the traditional bi-cubic interpolation and Yang’s method.

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

[2]  Yu-Chiang Frank Wang,et al.  Learning sparse image representation with support vector regression for single-image super-resolution , 2010, 2010 IEEE International Conference on Image Processing.

[3]  Thomas S. Huang,et al.  Image super-resolution as sparse representation of raw image patches , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Li-Wei Kang,et al.  Context-aware single image super-resolution using locality-constrained group sparse representation , 2012, 2012 Visual Communications and Image Processing.

[5]  Shutao Li,et al.  Infrared surveillance image super resolution via group sparse representation , 2013 .

[6]  Zhiyong Xu,et al.  Image super-resolution by dictionary concatenation and sparse representation with approximate L0 norm minimization , 2012, Comput. Electr. Eng..

[7]  John F. Roddick,et al.  Sparse representation-based MRI super-resolution reconstruction , 2014 .

[8]  Shutao Li,et al.  Simultaneous image fusion and super-resolution using sparse representation , 2013, Inf. Fusion.

[9]  Mahmoud Nazzal,et al.  Single image super resolution based on sparse representation via directionally structured dictionaries , 2014, 2014 22nd Signal Processing and Communications Applications Conference (SIU).

[10]  Shutao Li,et al.  Multi-morphology image super-resolution via sparse representation , 2013, Neurocomputing.

[11]  Di Zhang,et al.  Super-resolution image reconstruction via adaptive sparse representation and joint dictionary training , 2013, 2013 6th International Congress on Image and Signal Processing (CISP).

[12]  Shuyuan Yang,et al.  Multitask dictionary learning and sparse representation based single-image super-resolution reconstruction , 2011, Neurocomputing.

[13]  Jian Yang,et al.  A novel sparse representation based framework for face image super-resolution , 2014, Neurocomputing.

[14]  A. Bruckstein,et al.  K-SVD : An Algorithm for Designing of Overcomplete Dictionaries for Sparse Representation , 2005 .

[15]  Ruimin Hu,et al.  Robust super-resolution for face images via principle component sparse representation and least squares regression , 2013, 2013 IEEE International Symposium on Circuits and Systems (ISCAS2013).

[16]  Shutao Li,et al.  Single image super resolution via texture constrained sparse representation , 2011, 2011 18th IEEE International Conference on Image Processing.

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

[18]  Yunhui Shi,et al.  Single-Image Super-Resolution Based on Decomposition and Sparse Representation , 2010, 2010 International Conference on Multimedia Communications.

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