Single image super resolution based on sparse representation and adaptive dictionary selection

Chang Hong Fu; Hongli Chen; Hongbin Zhang; Yui Lam Chan

doi:10.1109/ICDSP.2014.6900704

Single image super resolution based on sparse representation and adaptive dictionary selection

Chang Hong Fu
, Hongli Chen
, Hongbin Zhang
, Yui Lam Chan

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

7 Citations (Scopus)

Abstract

An improved single image super resolution based on patch-wise sparse recovery is proposed in this paper. K-SVD is adopted to train a coupled dictionary. Besides, adaptive selection is proposed among dictionaries with different patch size. Simulation results show that the proposed approach provides good subjective quality and up to 0.4 dB PSNR improvement with significant time reduction.

Original language	English
Title of host publication	2014 19th International Conference on Digital Signal Processing, DSP 2014
Publisher	IEEE
Pages	449-453
Number of pages	5
Volume	2014-January
ISBN (Electronic)	9781479946129
DOIs	https://doi.org/10.1109/ICDSP.2014.6900704
Publication status	Published - 1 Jan 2014
Event	2014 19th International Conference on Digital Signal Processing, DSP 2014 - Hong Kong, Hong Kong Duration: 20 Aug 2014 → 23 Aug 2014

Conference

Conference	2014 19th International Conference on Digital Signal Processing, DSP 2014
Country/Territory	Hong Kong
City	Hong Kong
Period	20/08/14 → 23/08/14

Keywords

Adpative dictionary selection
Dictionary learning
K-svd
Sparse representation
Super resolution

ASJC Scopus subject areas

Signal Processing

More information

10.1109/ICDSP.2014.6900704

Cite this

@inproceedings{370e7cef084344ec88a0cafc9f0c8cba,

title = "Single image super resolution based on sparse representation and adaptive dictionary selection",

abstract = "An improved single image super resolution based on patch-wise sparse recovery is proposed in this paper. K-SVD is adopted to train a coupled dictionary. Besides, adaptive selection is proposed among dictionaries with different patch size. Simulation results show that the proposed approach provides good subjective quality and up to 0.4 dB PSNR improvement with significant time reduction.",

keywords = "Adpative dictionary selection, Dictionary learning, K-svd, Sparse representation, Super resolution",

author = "Fu, \{Chang Hong\} and Hongli Chen and Hongbin Zhang and Chan, \{Yui Lam\}",

year = "2014",

month = jan,

day = "1",

doi = "10.1109/ICDSP.2014.6900704",

language = "English",

volume = "2014-January",

pages = "449--453",

booktitle = "2014 19th International Conference on Digital Signal Processing, DSP 2014",

publisher = "IEEE",

address = "United States",

note = "2014 19th International Conference on Digital Signal Processing, DSP 2014 ; Conference date: 20-08-2014 Through 23-08-2014",

}

Fu, CH, Chen, H, Zhang, H & Chan, YL 2014, Single image super resolution based on sparse representation and adaptive dictionary selection. in 2014 19th International Conference on Digital Signal Processing, DSP 2014. vol. 2014-January, 6900704, IEEE, pp. 449-453, 2014 19th International Conference on Digital Signal Processing, DSP 2014, Hong Kong, Hong Kong, 20/08/14. https://doi.org/10.1109/ICDSP.2014.6900704

Single image super resolution based on sparse representation and adaptive dictionary selection. / Fu, Chang Hong; Chen, Hongli; Zhang, Hongbin et al.
2014 19th International Conference on Digital Signal Processing, DSP 2014. Vol. 2014-January IEEE, 2014. p. 449-453 6900704.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Single image super resolution based on sparse representation and adaptive dictionary selection

AU - Fu, Chang Hong

AU - Chen, Hongli

AU - Zhang, Hongbin

AU - Chan, Yui Lam

PY - 2014/1/1

Y1 - 2014/1/1

N2 - An improved single image super resolution based on patch-wise sparse recovery is proposed in this paper. K-SVD is adopted to train a coupled dictionary. Besides, adaptive selection is proposed among dictionaries with different patch size. Simulation results show that the proposed approach provides good subjective quality and up to 0.4 dB PSNR improvement with significant time reduction.

AB - An improved single image super resolution based on patch-wise sparse recovery is proposed in this paper. K-SVD is adopted to train a coupled dictionary. Besides, adaptive selection is proposed among dictionaries with different patch size. Simulation results show that the proposed approach provides good subjective quality and up to 0.4 dB PSNR improvement with significant time reduction.

KW - Adpative dictionary selection

KW - Dictionary learning

KW - K-svd

KW - Sparse representation

KW - Super resolution

UR - https://www.scopus.com/pages/publications/84940761233

U2 - 10.1109/ICDSP.2014.6900704

DO - 10.1109/ICDSP.2014.6900704

M3 - Conference article published in proceeding or book

VL - 2014-January

SP - 449

EP - 453

BT - 2014 19th International Conference on Digital Signal Processing, DSP 2014

PB - IEEE

T2 - 2014 19th International Conference on Digital Signal Processing, DSP 2014

Y2 - 20 August 2014 through 23 August 2014

ER -

Single image super resolution based on sparse representation and adaptive dictionary selection

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this