Image reconstruction using the Viterbi algorithm

C. Miller, B. R. Hunt, Mark A Neifeld, Michael W Marcellin

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Many systems in widespread use concentrate on the imaging of binary objects, e.g., the archival storage of text documents on microfilm or the facsimile transmission of text. Due to the imperfect nature of such systems, the binary image is unavoidably corrupted by blur and noise to form a grey-scale image. We present a technique to reverse this degradation which maps the binary object reconstruction problem into a Viterbi state-trellis. We assign states of the trellis to possible outcomes of the reconstruction estimate and search the trellis in the usual optimal fashion. Our method yields superior estimates of the original binary object over a wide range of signal-to-noise ratios (SNR) when compared with conventional Wiener filter (WF) estimates. For moderate blur and SNR levels, the estimates produced approach the maximum likelihood (ML) bound on estimation performance.

Original languageEnglish (US)
Title of host publicationProceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation
PublisherIEEE
Pages160-163
Number of pages4
StatePublished - 1998
EventProceedings of the 1998 IEEE Southwest Symposium on Image Analysis and Interpretation - Tucson, AZ, USA
Duration: Apr 5 1998Apr 7 1998

Other

OtherProceedings of the 1998 IEEE Southwest Symposium on Image Analysis and Interpretation
CityTucson, AZ, USA
Period4/5/984/7/98

Fingerprint

Viterbi algorithm
Image reconstruction
Signal to noise ratio
Microfilm
Facsimile
Binary images
Maximum likelihood
Imaging techniques
Degradation

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Miller, C., Hunt, B. R., Neifeld, M. A., & Marcellin, M. W. (1998). Image reconstruction using the Viterbi algorithm. In Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation (pp. 160-163). IEEE.

Image reconstruction using the Viterbi algorithm. / Miller, C.; Hunt, B. R.; Neifeld, Mark A; Marcellin, Michael W.

Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation. IEEE, 1998. p. 160-163.

Research output: Chapter in Book/Report/Conference proceedingChapter

Miller, C, Hunt, BR, Neifeld, MA & Marcellin, MW 1998, Image reconstruction using the Viterbi algorithm. in Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation. IEEE, pp. 160-163, Proceedings of the 1998 IEEE Southwest Symposium on Image Analysis and Interpretation, Tucson, AZ, USA, 4/5/98.
Miller C, Hunt BR, Neifeld MA, Marcellin MW. Image reconstruction using the Viterbi algorithm. In Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation. IEEE. 1998. p. 160-163
Miller, C. ; Hunt, B. R. ; Neifeld, Mark A ; Marcellin, Michael W. / Image reconstruction using the Viterbi algorithm. Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation. IEEE, 1998. pp. 160-163
@inbook{9e140a12f6b14bc183a2fd6e953ccfeb,
title = "Image reconstruction using the Viterbi algorithm",
abstract = "Many systems in widespread use concentrate on the imaging of binary objects, e.g., the archival storage of text documents on microfilm or the facsimile transmission of text. Due to the imperfect nature of such systems, the binary image is unavoidably corrupted by blur and noise to form a grey-scale image. We present a technique to reverse this degradation which maps the binary object reconstruction problem into a Viterbi state-trellis. We assign states of the trellis to possible outcomes of the reconstruction estimate and search the trellis in the usual optimal fashion. Our method yields superior estimates of the original binary object over a wide range of signal-to-noise ratios (SNR) when compared with conventional Wiener filter (WF) estimates. For moderate blur and SNR levels, the estimates produced approach the maximum likelihood (ML) bound on estimation performance.",
author = "C. Miller and Hunt, {B. R.} and Neifeld, {Mark A} and Marcellin, {Michael W}",
year = "1998",
language = "English (US)",
pages = "160--163",
booktitle = "Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation",
publisher = "IEEE",

}

TY - CHAP

T1 - Image reconstruction using the Viterbi algorithm

AU - Miller, C.

AU - Hunt, B. R.

AU - Neifeld, Mark A

AU - Marcellin, Michael W

PY - 1998

Y1 - 1998

N2 - Many systems in widespread use concentrate on the imaging of binary objects, e.g., the archival storage of text documents on microfilm or the facsimile transmission of text. Due to the imperfect nature of such systems, the binary image is unavoidably corrupted by blur and noise to form a grey-scale image. We present a technique to reverse this degradation which maps the binary object reconstruction problem into a Viterbi state-trellis. We assign states of the trellis to possible outcomes of the reconstruction estimate and search the trellis in the usual optimal fashion. Our method yields superior estimates of the original binary object over a wide range of signal-to-noise ratios (SNR) when compared with conventional Wiener filter (WF) estimates. For moderate blur and SNR levels, the estimates produced approach the maximum likelihood (ML) bound on estimation performance.

AB - Many systems in widespread use concentrate on the imaging of binary objects, e.g., the archival storage of text documents on microfilm or the facsimile transmission of text. Due to the imperfect nature of such systems, the binary image is unavoidably corrupted by blur and noise to form a grey-scale image. We present a technique to reverse this degradation which maps the binary object reconstruction problem into a Viterbi state-trellis. We assign states of the trellis to possible outcomes of the reconstruction estimate and search the trellis in the usual optimal fashion. Our method yields superior estimates of the original binary object over a wide range of signal-to-noise ratios (SNR) when compared with conventional Wiener filter (WF) estimates. For moderate blur and SNR levels, the estimates produced approach the maximum likelihood (ML) bound on estimation performance.

UR - http://www.scopus.com/inward/record.url?scp=0032230835&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032230835&partnerID=8YFLogxK

M3 - Chapter

SP - 160

EP - 163

BT - Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation

PB - IEEE

ER -