Super-resolution reconstruction using wavelet transform

Toshio Ito, Hiroki Kobayashi, Taiki Sekii, Takenao Ohkawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Because of a deterioration in resolution, the distant parts of images taken by acamera are of an inferior quality. Super-resolution is one of the effective methods to improvethe quality of low resolution images. In super-resolution, a high resolution-image is generatedfrom multiple low resolution images by image processing such as inverse discrete wavelettransformation. Inverse discrete wavelet transform can expand the image by doubling theresolution. Super-resolution by inverse discrete wavelet transform needs three different highfrequency components in the image to be improved. For accurate expansion by inversediscrete wavelet transform, these high frequency components must be accurate. However, it isdifficult to estimate these components because of the sensitivity of the transform toward themargin of error. This is why there are few methods for improving images by using wavelettransform. Therefore, we have tried to reduce this domain to estimate. Estimation of highfrequency components becomes easy, if we can improve low resolution images after thedomain has been reduced. In this paper, we have tried to limit the domain of these highfrequency components.

Original languageEnglish
Title of host publication15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008
Pages5817-5828
Number of pages12
Volume8
Publication statusPublished - 2008
Externally publishedYes
Event15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008 - New York, NY
Duration: 2008 Nov 162008 Nov 20

Other

Other15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008
CityNew York, NY
Period08/11/1608/11/20

Fingerprint

Image resolution
Wavelet transforms
reconstruction
Discrete wavelet transforms
Optical resolving power
Deterioration
Image processing

Keywords

  • MRA
  • On-board camera
  • Super-resolution
  • Vehicle recognition
  • Wavelet transform

ASJC Scopus subject areas

  • Mechanical Engineering
  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Transportation
  • Automotive Engineering
  • Computer Networks and Communications
  • Artificial Intelligence
  • Computer Science Applications

Cite this

Ito, T., Kobayashi, H., Sekii, T., & Ohkawa, T. (2008). Super-resolution reconstruction using wavelet transform. In 15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008 (Vol. 8, pp. 5817-5828)

Super-resolution reconstruction using wavelet transform. / Ito, Toshio; Kobayashi, Hiroki; Sekii, Taiki; Ohkawa, Takenao.

15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008. Vol. 8 2008. p. 5817-5828.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ito, T, Kobayashi, H, Sekii, T & Ohkawa, T 2008, Super-resolution reconstruction using wavelet transform. in 15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008. vol. 8, pp. 5817-5828, 15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008, New York, NY, 08/11/16.
Ito T, Kobayashi H, Sekii T, Ohkawa T. Super-resolution reconstruction using wavelet transform. In 15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008. Vol. 8. 2008. p. 5817-5828
Ito, Toshio ; Kobayashi, Hiroki ; Sekii, Taiki ; Ohkawa, Takenao. / Super-resolution reconstruction using wavelet transform. 15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008. Vol. 8 2008. pp. 5817-5828
@inproceedings{b7434e21e93f43be8240da1d8cb3f336,
title = "Super-resolution reconstruction using wavelet transform",
abstract = "Because of a deterioration in resolution, the distant parts of images taken by acamera are of an inferior quality. Super-resolution is one of the effective methods to improvethe quality of low resolution images. In super-resolution, a high resolution-image is generatedfrom multiple low resolution images by image processing such as inverse discrete wavelettransformation. Inverse discrete wavelet transform can expand the image by doubling theresolution. Super-resolution by inverse discrete wavelet transform needs three different highfrequency components in the image to be improved. For accurate expansion by inversediscrete wavelet transform, these high frequency components must be accurate. However, it isdifficult to estimate these components because of the sensitivity of the transform toward themargin of error. This is why there are few methods for improving images by using wavelettransform. Therefore, we have tried to reduce this domain to estimate. Estimation of highfrequency components becomes easy, if we can improve low resolution images after thedomain has been reduced. In this paper, we have tried to limit the domain of these highfrequency components.",
keywords = "MRA, On-board camera, Super-resolution, Vehicle recognition, Wavelet transform",
author = "Toshio Ito and Hiroki Kobayashi and Taiki Sekii and Takenao Ohkawa",
year = "2008",
language = "English",
isbn = "9781615677566",
volume = "8",
pages = "5817--5828",
booktitle = "15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008",

}

TY - GEN

T1 - Super-resolution reconstruction using wavelet transform

AU - Ito, Toshio

AU - Kobayashi, Hiroki

AU - Sekii, Taiki

AU - Ohkawa, Takenao

PY - 2008

Y1 - 2008

N2 - Because of a deterioration in resolution, the distant parts of images taken by acamera are of an inferior quality. Super-resolution is one of the effective methods to improvethe quality of low resolution images. In super-resolution, a high resolution-image is generatedfrom multiple low resolution images by image processing such as inverse discrete wavelettransformation. Inverse discrete wavelet transform can expand the image by doubling theresolution. Super-resolution by inverse discrete wavelet transform needs three different highfrequency components in the image to be improved. For accurate expansion by inversediscrete wavelet transform, these high frequency components must be accurate. However, it isdifficult to estimate these components because of the sensitivity of the transform toward themargin of error. This is why there are few methods for improving images by using wavelettransform. Therefore, we have tried to reduce this domain to estimate. Estimation of highfrequency components becomes easy, if we can improve low resolution images after thedomain has been reduced. In this paper, we have tried to limit the domain of these highfrequency components.

AB - Because of a deterioration in resolution, the distant parts of images taken by acamera are of an inferior quality. Super-resolution is one of the effective methods to improvethe quality of low resolution images. In super-resolution, a high resolution-image is generatedfrom multiple low resolution images by image processing such as inverse discrete wavelettransformation. Inverse discrete wavelet transform can expand the image by doubling theresolution. Super-resolution by inverse discrete wavelet transform needs three different highfrequency components in the image to be improved. For accurate expansion by inversediscrete wavelet transform, these high frequency components must be accurate. However, it isdifficult to estimate these components because of the sensitivity of the transform toward themargin of error. This is why there are few methods for improving images by using wavelettransform. Therefore, we have tried to reduce this domain to estimate. Estimation of highfrequency components becomes easy, if we can improve low resolution images after thedomain has been reduced. In this paper, we have tried to limit the domain of these highfrequency components.

KW - MRA

KW - On-board camera

KW - Super-resolution

KW - Vehicle recognition

KW - Wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=84878918685&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878918685&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781615677566

VL - 8

SP - 5817

EP - 5828

BT - 15th World Congress on Intelligent Transport Systems and ITS America Annual Meeting 2008

ER -