Super-resolution algorithm based on weighted vector quantization

Tse Ming Kuo, Shen-Chuan Tai

Research output: Contribution to journalArticle

Abstract

Super resolution is a technique that obtains high-resolution (HR) images from the corresponding lowresolution (LR) ones and moreover, makes these HR images as natural as possible. There are usually some artifacts of these HR images caused by the lack of the information of the detailed textures (high-frequency information). In our research, an algorithm based on weighted vector quantization is proposed to predict the highfrequency information. First of all, the classified vector quantization is adopted to establish a pair of codebooks of high-frequency information for LR and HR patches. Second, three code vectors are selected from the LR codebooks based on the correlation coefficients to establish a multilinear regression model with the input LR patch and meanwhile the weights are decided adaptively. Then, three corresponding HR code vectors are used with the weights to reconstruct the high-frequency information of the HR patch. Finally, the reconstructed high-frequency information is used to refine the preliminarily upscaled HR image. The experimental results show that the performance of the proposed algorithm is good not only in objective but also in subjective measurement.

Original languageEnglish
Article number023002
JournalJournal of Electronic Imaging
Volume23
Issue number2
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

vector quantization
Vector quantization
Image resolution
high resolution
Optical resolving power
Textures
correlation coefficients
artifacts
regression analysis
textures

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

@article{4bca04f2b4bf432fbca0f036f2105542,
title = "Super-resolution algorithm based on weighted vector quantization",
abstract = "Super resolution is a technique that obtains high-resolution (HR) images from the corresponding lowresolution (LR) ones and moreover, makes these HR images as natural as possible. There are usually some artifacts of these HR images caused by the lack of the information of the detailed textures (high-frequency information). In our research, an algorithm based on weighted vector quantization is proposed to predict the highfrequency information. First of all, the classified vector quantization is adopted to establish a pair of codebooks of high-frequency information for LR and HR patches. Second, three code vectors are selected from the LR codebooks based on the correlation coefficients to establish a multilinear regression model with the input LR patch and meanwhile the weights are decided adaptively. Then, three corresponding HR code vectors are used with the weights to reconstruct the high-frequency information of the HR patch. Finally, the reconstructed high-frequency information is used to refine the preliminarily upscaled HR image. The experimental results show that the performance of the proposed algorithm is good not only in objective but also in subjective measurement.",
author = "Kuo, {Tse Ming} and Shen-Chuan Tai",
year = "2014",
month = "1",
day = "1",
doi = "10.1117/1.JEI.23.2.023002",
language = "English",
volume = "23",
journal = "Journal of Electronic Imaging",
issn = "1017-9909",
publisher = "SPIE",
number = "2",

}

Super-resolution algorithm based on weighted vector quantization. / Kuo, Tse Ming; Tai, Shen-Chuan.

In: Journal of Electronic Imaging, Vol. 23, No. 2, 023002, 01.01.2014.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Super-resolution algorithm based on weighted vector quantization

AU - Kuo, Tse Ming

AU - Tai, Shen-Chuan

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Super resolution is a technique that obtains high-resolution (HR) images from the corresponding lowresolution (LR) ones and moreover, makes these HR images as natural as possible. There are usually some artifacts of these HR images caused by the lack of the information of the detailed textures (high-frequency information). In our research, an algorithm based on weighted vector quantization is proposed to predict the highfrequency information. First of all, the classified vector quantization is adopted to establish a pair of codebooks of high-frequency information for LR and HR patches. Second, three code vectors are selected from the LR codebooks based on the correlation coefficients to establish a multilinear regression model with the input LR patch and meanwhile the weights are decided adaptively. Then, three corresponding HR code vectors are used with the weights to reconstruct the high-frequency information of the HR patch. Finally, the reconstructed high-frequency information is used to refine the preliminarily upscaled HR image. The experimental results show that the performance of the proposed algorithm is good not only in objective but also in subjective measurement.

AB - Super resolution is a technique that obtains high-resolution (HR) images from the corresponding lowresolution (LR) ones and moreover, makes these HR images as natural as possible. There are usually some artifacts of these HR images caused by the lack of the information of the detailed textures (high-frequency information). In our research, an algorithm based on weighted vector quantization is proposed to predict the highfrequency information. First of all, the classified vector quantization is adopted to establish a pair of codebooks of high-frequency information for LR and HR patches. Second, three code vectors are selected from the LR codebooks based on the correlation coefficients to establish a multilinear regression model with the input LR patch and meanwhile the weights are decided adaptively. Then, three corresponding HR code vectors are used with the weights to reconstruct the high-frequency information of the HR patch. Finally, the reconstructed high-frequency information is used to refine the preliminarily upscaled HR image. The experimental results show that the performance of the proposed algorithm is good not only in objective but also in subjective measurement.

UR - http://www.scopus.com/inward/record.url?scp=84896365482&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896365482&partnerID=8YFLogxK

U2 - 10.1117/1.JEI.23.2.023002

DO - 10.1117/1.JEI.23.2.023002

M3 - Article

VL - 23

JO - Journal of Electronic Imaging

JF - Journal of Electronic Imaging

SN - 1017-9909

IS - 2

M1 - 023002

ER -