Sparse Representation for Color Image Super-Resolution with Image Quality Difference Evaluation

Zi-wen WANG; Guo-rui FENG; Ling-yan FAN; Jin-wei WANG

doi:10.1587/transinf.2016EDP7217

Sparse Representation for Color Image Super-Resolution with Image Quality Difference Evaluation

Zi-wen WANG, Guo-rui FENG, Ling-yan FAN, Jin-wei WANG

Full Text Views

0

Share
Cite this

Summary :

The sparse representation models have been widely applied in image super-resolution. The certain optimization problem is supposed and can be solved by the iterative shrinkage algorithm. During iteration, the update of dictionaries and similar patches is necessary to obtain prior knowledge to better solve such ill-conditioned problem as image super-resolution. However, both the processes of iteration and update often spend a lot of time, which will be a bottleneck in practice. To solve it, in this paper, we present the concept of image quality difference based on generalized Gaussian distribution feature which has the same trend with the variation of Peak Signal to Noise Ratio (PSNR), and we update dictionaries or similar patches from the termination strategy according to the adaptive threshold of the image quality difference. Based on this point, we present two sparse representation algorithms for image super-resolution, one achieves the further improvement in image quality and the other decreases running time on the basis of image quality assurance. Experimental results also show that our quantitative results on several test datasets are in line with exceptions.

Publication: IEICE TRANSACTIONS on Information Vol.E100-D No.1 pp.150-159

Publication Date: 2017/01/01

Publicized: 2016/10/19

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2016EDP7217

Type of Manuscript: PAPER

Category: Image Processing and Video Processing

Authors

Zi-wen WANG
  Shanghai University
Guo-rui FENG
  Shanghai University
Ling-yan FAN
  Hangzhou Dianzi University
Jin-wei WANG
  Nanjing University of Information Science & Technology

Keyword

image super-resolution, sparse representation, generalized Gaussian distribution feature, image quality difference, fast iterative shrinkage algorithm

Cite this

Copy

Zi-wen WANG, Guo-rui FENG, Ling-yan FAN, Jin-wei WANG, "Sparse Representation for Color Image Super-Resolution with Image Quality Difference Evaluation" in IEICE TRANSACTIONS on Information, vol. E100-D, no. 1, pp. 150-159, January 2017, doi: 10.1587/transinf.2016EDP7217.
Abstract: The sparse representation models have been widely applied in image super-resolution. The certain optimization problem is supposed and can be solved by the iterative shrinkage algorithm. During iteration, the update of dictionaries and similar patches is necessary to obtain prior knowledge to better solve such ill-conditioned problem as image super-resolution. However, both the processes of iteration and update often spend a lot of time, which will be a bottleneck in practice. To solve it, in this paper, we present the concept of image quality difference based on generalized Gaussian distribution feature which has the same trend with the variation of Peak Signal to Noise Ratio (PSNR), and we update dictionaries or similar patches from the termination strategy according to the adaptive threshold of the image quality difference. Based on this point, we present two sparse representation algorithms for image super-resolution, one achieves the further improvement in image quality and the other decreases running time on the basis of image quality assurance. Experimental results also show that our quantitative results on several test datasets are in line with exceptions.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.2016EDP7217/_p

Copy

@ARTICLE{e100-d_1_150,
author={Zi-wen WANG, Guo-rui FENG, Ling-yan FAN, Jin-wei WANG, },
journal={IEICE TRANSACTIONS on Information},
title={Sparse Representation for Color Image Super-Resolution with Image Quality Difference Evaluation},
year={2017},
volume={E100-D},
number={1},
pages={150-159},
abstract={The sparse representation models have been widely applied in image super-resolution. The certain optimization problem is supposed and can be solved by the iterative shrinkage algorithm. During iteration, the update of dictionaries and similar patches is necessary to obtain prior knowledge to better solve such ill-conditioned problem as image super-resolution. However, both the processes of iteration and update often spend a lot of time, which will be a bottleneck in practice. To solve it, in this paper, we present the concept of image quality difference based on generalized Gaussian distribution feature which has the same trend with the variation of Peak Signal to Noise Ratio (PSNR), and we update dictionaries or similar patches from the termination strategy according to the adaptive threshold of the image quality difference. Based on this point, we present two sparse representation algorithms for image super-resolution, one achieves the further improvement in image quality and the other decreases running time on the basis of image quality assurance. Experimental results also show that our quantitative results on several test datasets are in line with exceptions.},
keywords={},
doi={10.1587/transinf.2016EDP7217},
ISSN={1745-1361},
month={January},}

Copy

TY - JOUR
TI - Sparse Representation for Color Image Super-Resolution with Image Quality Difference Evaluation
T2 - IEICE TRANSACTIONS on Information
SP - 150
EP - 159
AU - Zi-wen WANG
AU - Guo-rui FENG
AU - Ling-yan FAN
AU - Jin-wei WANG
PY - 2017
DO - 10.1587/transinf.2016EDP7217
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2017
AB - The sparse representation models have been widely applied in image super-resolution. The certain optimization problem is supposed and can be solved by the iterative shrinkage algorithm. During iteration, the update of dictionaries and similar patches is necessary to obtain prior knowledge to better solve such ill-conditioned problem as image super-resolution. However, both the processes of iteration and update often spend a lot of time, which will be a bottleneck in practice. To solve it, in this paper, we present the concept of image quality difference based on generalized Gaussian distribution feature which has the same trend with the variation of Peak Signal to Noise Ratio (PSNR), and we update dictionaries or similar patches from the termination strategy according to the adaptive threshold of the image quality difference. Based on this point, we present two sparse representation algorithms for image super-resolution, one achieves the further improvement in image quality and the other decreases running time on the basis of image quality assurance. Experimental results also show that our quantitative results on several test datasets are in line with exceptions.
ER -