Improved Interpolation Kernels for Super-resolution Algorithms

Pejman Rasti, Olga Orlova, Gert Tamberg, Cagri Ozcinar, Kamal Nasrollahi, Thomas B. Moeslund, Gholamreza Anbarjafari

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

6 Citations (Scopus)

Abstract

Super resolution (SR) algorithms are widely used in forensics investigations to enhance the resolution of images captured by surveillance cameras. Such algorithms usually use a common interpolation algorithm to generate an initial guess for the desired high resolution (HR) image. This initial guess is usually tuned through different methods, like learning-based or fusion-based methods, to converge the initial guess towards the desired HR output. In this work, it is shown that SR algorithms can result in better performance if more sophisticated kernels than the simple conventional ones are used for producing the initial guess. The contribution of this work is to introduce such a set of kernels which can be used in the context of SR. The quantitative and qualitative results on many natural, facial and iris images show the superiority of the generated HR images over two state-of-the-art SR algorithms when their original interpolation kernel is replaced by the ones introduced in this work.
Original languageEnglish
Title of host publicationIEEE International Conference on Image Processing Theory, Tools and Applications
PublisherIEEE
Publication dateDec 2016
ISBN (Print)978-1-4673-8911-2
ISBN (Electronic)978-1-4673-8910-5
DOIs
Publication statusPublished - Dec 2016
EventIEEE International Conference on Image Processing Theory, Tools and Applications - Oulu, Finland
Duration: 12 Dec 201615 Dec 2016
Conference number: 6
http://www.ipta-conference.com/ipta16/

Conference

ConferenceIEEE International Conference on Image Processing Theory, Tools and Applications
Number6
CountryFinland
CityOulu
Period12/12/201615/12/2016
Internet address

Fingerprint

Interpolation
Image resolution
Fusion reactions
Cameras

Keywords

  • Sampling kernels
  • resolution enhancement
  • upsampling
  • sampling theory

Cite this

Rasti, P., Orlova, O., Tamberg, G., Ozcinar, C., Nasrollahi, K., Moeslund, T. B., & Anbarjafari, G. (2016). Improved Interpolation Kernels for Super-resolution Algorithms. In IEEE International Conference on Image Processing Theory, Tools and Applications IEEE. https://doi.org/10.1109/IPTA.2016.7820980
Rasti, Pejman ; Orlova, Olga ; Tamberg, Gert ; Ozcinar, Cagri ; Nasrollahi, Kamal ; Moeslund, Thomas B. ; Anbarjafari, Gholamreza. / Improved Interpolation Kernels for Super-resolution Algorithms. IEEE International Conference on Image Processing Theory, Tools and Applications. IEEE, 2016.
@inproceedings{db2f9f6ea49540dda9f2559bf0446b45,
title = "Improved Interpolation Kernels for Super-resolution Algorithms",
abstract = "Super resolution (SR) algorithms are widely used in forensics investigations to enhance the resolution of images captured by surveillance cameras. Such algorithms usually use a common interpolation algorithm to generate an initial guess for the desired high resolution (HR) image. This initial guess is usually tuned through different methods, like learning-based or fusion-based methods, to converge the initial guess towards the desired HR output. In this work, it is shown that SR algorithms can result in better performance if more sophisticated kernels than the simple conventional ones are used for producing the initial guess. The contribution of this work is to introduce such a set of kernels which can be used in the context of SR. The quantitative and qualitative results on many natural, facial and iris images show the superiority of the generated HR images over two state-of-the-art SR algorithms when their original interpolation kernel is replaced by the ones introduced in this work.",
keywords = "Sampling kernels, resolution enhancement, upsampling, sampling theory",
author = "Pejman Rasti and Olga Orlova and Gert Tamberg and Cagri Ozcinar and Kamal Nasrollahi and Moeslund, {Thomas B.} and Gholamreza Anbarjafari",
year = "2016",
month = "12",
doi = "10.1109/IPTA.2016.7820980",
language = "English",
isbn = "978-1-4673-8911-2",
booktitle = "IEEE International Conference on Image Processing Theory, Tools and Applications",
publisher = "IEEE",
address = "United States",

}

Rasti, P, Orlova, O, Tamberg, G, Ozcinar, C, Nasrollahi, K, Moeslund, TB & Anbarjafari, G 2016, Improved Interpolation Kernels for Super-resolution Algorithms. in IEEE International Conference on Image Processing Theory, Tools and Applications. IEEE, IEEE International Conference on Image Processing Theory, Tools and Applications, Oulu, Finland, 12/12/2016. https://doi.org/10.1109/IPTA.2016.7820980

Improved Interpolation Kernels for Super-resolution Algorithms. / Rasti, Pejman; Orlova, Olga; Tamberg, Gert; Ozcinar, Cagri; Nasrollahi, Kamal; Moeslund, Thomas B.; Anbarjafari, Gholamreza.

IEEE International Conference on Image Processing Theory, Tools and Applications. IEEE, 2016.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Improved Interpolation Kernels for Super-resolution Algorithms

AU - Rasti, Pejman

AU - Orlova, Olga

AU - Tamberg, Gert

AU - Ozcinar, Cagri

AU - Nasrollahi, Kamal

AU - Moeslund, Thomas B.

AU - Anbarjafari, Gholamreza

PY - 2016/12

Y1 - 2016/12

N2 - Super resolution (SR) algorithms are widely used in forensics investigations to enhance the resolution of images captured by surveillance cameras. Such algorithms usually use a common interpolation algorithm to generate an initial guess for the desired high resolution (HR) image. This initial guess is usually tuned through different methods, like learning-based or fusion-based methods, to converge the initial guess towards the desired HR output. In this work, it is shown that SR algorithms can result in better performance if more sophisticated kernels than the simple conventional ones are used for producing the initial guess. The contribution of this work is to introduce such a set of kernels which can be used in the context of SR. The quantitative and qualitative results on many natural, facial and iris images show the superiority of the generated HR images over two state-of-the-art SR algorithms when their original interpolation kernel is replaced by the ones introduced in this work.

AB - Super resolution (SR) algorithms are widely used in forensics investigations to enhance the resolution of images captured by surveillance cameras. Such algorithms usually use a common interpolation algorithm to generate an initial guess for the desired high resolution (HR) image. This initial guess is usually tuned through different methods, like learning-based or fusion-based methods, to converge the initial guess towards the desired HR output. In this work, it is shown that SR algorithms can result in better performance if more sophisticated kernels than the simple conventional ones are used for producing the initial guess. The contribution of this work is to introduce such a set of kernels which can be used in the context of SR. The quantitative and qualitative results on many natural, facial and iris images show the superiority of the generated HR images over two state-of-the-art SR algorithms when their original interpolation kernel is replaced by the ones introduced in this work.

KW - Sampling kernels

KW - resolution enhancement

KW - upsampling

KW - sampling theory

UR - http://www.ipta-conference.com/ipta16/images/IPTA_program_Final_Online.pdf

U2 - 10.1109/IPTA.2016.7820980

DO - 10.1109/IPTA.2016.7820980

M3 - Article in proceeding

SN - 978-1-4673-8911-2

BT - IEEE International Conference on Image Processing Theory, Tools and Applications

PB - IEEE

ER -

Rasti P, Orlova O, Tamberg G, Ozcinar C, Nasrollahi K, Moeslund TB et al. Improved Interpolation Kernels for Super-resolution Algorithms. In IEEE International Conference on Image Processing Theory, Tools and Applications. IEEE. 2016 https://doi.org/10.1109/IPTA.2016.7820980