Regularized high-resolution image reconstruction considering inaccurate motion information

Min Kyu Park, Moon Gi Kang, Aggelos K. Katsaggelos

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

One of the major issues in recovering a high-resolution image from a sequence of low-resolution observations is the accuracy of the motion information. In most of the work in the literature, the motion information is assumed to be known with high accuracy. This is very often not the case, and therefore the accuracy of high-resolution image reconstruction suffers substantially, since it greatly depends on the motion information. To address these issues in this paper, we propose a high-resolution image reconstruction algorithm that reduces the distortion in the reconstructed high-resolution image due to the inaccuracy of the estimated motion. Towards this task, we analyze the reconstruction noise generated by the inaccurate motion information. Based on this analysis, we propose a new regularization functional and derive a sufficient condition for the convergence of the resulting iterative reconstruction algorithm. The proposed algorithm requires no prior information about the original image or the inaccuracy of the motion information. Experimental results illustrate the benefit of the proposed method when compared to conventional high-resolution image reconstruction methods in terms of both objective measurements and subjective evaluation.

Original languageEnglish
Article number117004
JournalOptical Engineering
Volume46
Issue number11
DOIs
Publication statusPublished - 2007 Dec 10

Fingerprint

image reconstruction
Image reconstruction
Image resolution
high resolution
Optical resolving power
evaluation

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics
  • Engineering(all)

Cite this

@article{1c7cc027a56a4fe4a2cb52ad9ae54f1f,
title = "Regularized high-resolution image reconstruction considering inaccurate motion information",
abstract = "One of the major issues in recovering a high-resolution image from a sequence of low-resolution observations is the accuracy of the motion information. In most of the work in the literature, the motion information is assumed to be known with high accuracy. This is very often not the case, and therefore the accuracy of high-resolution image reconstruction suffers substantially, since it greatly depends on the motion information. To address these issues in this paper, we propose a high-resolution image reconstruction algorithm that reduces the distortion in the reconstructed high-resolution image due to the inaccuracy of the estimated motion. Towards this task, we analyze the reconstruction noise generated by the inaccurate motion information. Based on this analysis, we propose a new regularization functional and derive a sufficient condition for the convergence of the resulting iterative reconstruction algorithm. The proposed algorithm requires no prior information about the original image or the inaccuracy of the motion information. Experimental results illustrate the benefit of the proposed method when compared to conventional high-resolution image reconstruction methods in terms of both objective measurements and subjective evaluation.",
author = "Park, {Min Kyu} and Kang, {Moon Gi} and Katsaggelos, {Aggelos K.}",
year = "2007",
month = "12",
day = "10",
doi = "10.1117/1.2802611",
language = "English",
volume = "46",
journal = "Optical Engineering",
issn = "0091-3286",
publisher = "SPIE",
number = "11",

}

Regularized high-resolution image reconstruction considering inaccurate motion information. / Park, Min Kyu; Kang, Moon Gi; Katsaggelos, Aggelos K.

In: Optical Engineering, Vol. 46, No. 11, 117004, 10.12.2007.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Regularized high-resolution image reconstruction considering inaccurate motion information

AU - Park, Min Kyu

AU - Kang, Moon Gi

AU - Katsaggelos, Aggelos K.

PY - 2007/12/10

Y1 - 2007/12/10

N2 - One of the major issues in recovering a high-resolution image from a sequence of low-resolution observations is the accuracy of the motion information. In most of the work in the literature, the motion information is assumed to be known with high accuracy. This is very often not the case, and therefore the accuracy of high-resolution image reconstruction suffers substantially, since it greatly depends on the motion information. To address these issues in this paper, we propose a high-resolution image reconstruction algorithm that reduces the distortion in the reconstructed high-resolution image due to the inaccuracy of the estimated motion. Towards this task, we analyze the reconstruction noise generated by the inaccurate motion information. Based on this analysis, we propose a new regularization functional and derive a sufficient condition for the convergence of the resulting iterative reconstruction algorithm. The proposed algorithm requires no prior information about the original image or the inaccuracy of the motion information. Experimental results illustrate the benefit of the proposed method when compared to conventional high-resolution image reconstruction methods in terms of both objective measurements and subjective evaluation.

AB - One of the major issues in recovering a high-resolution image from a sequence of low-resolution observations is the accuracy of the motion information. In most of the work in the literature, the motion information is assumed to be known with high accuracy. This is very often not the case, and therefore the accuracy of high-resolution image reconstruction suffers substantially, since it greatly depends on the motion information. To address these issues in this paper, we propose a high-resolution image reconstruction algorithm that reduces the distortion in the reconstructed high-resolution image due to the inaccuracy of the estimated motion. Towards this task, we analyze the reconstruction noise generated by the inaccurate motion information. Based on this analysis, we propose a new regularization functional and derive a sufficient condition for the convergence of the resulting iterative reconstruction algorithm. The proposed algorithm requires no prior information about the original image or the inaccuracy of the motion information. Experimental results illustrate the benefit of the proposed method when compared to conventional high-resolution image reconstruction methods in terms of both objective measurements and subjective evaluation.

UR - http://www.scopus.com/inward/record.url?scp=36749099930&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36749099930&partnerID=8YFLogxK

U2 - 10.1117/1.2802611

DO - 10.1117/1.2802611

M3 - Article

VL - 46

JO - Optical Engineering

JF - Optical Engineering

SN - 0091-3286

IS - 11

M1 - 117004

ER -