Binary image reconstruction via 2-D Viterbi search

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

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

27 Citations (Scopus)

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 publicationIEEE International Conference on Image Processing
PublisherIEEE Comp Soc
Pages181-184
Number of pages4
Volume1
StatePublished - 1997
EventProceedings of the 1997 International Conference on Image Processing. Part 2 (of 3) - Santa Barbara, CA, USA
Duration: Oct 26 1997Oct 29 1997

Other

OtherProceedings of the 1997 International Conference on Image Processing. Part 2 (of 3)
CitySanta Barbara, CA, USA
Period10/26/9710/29/97

Fingerprint

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

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Miller, C., Hunt, B. R., Neifeld, M. A., & Marcellin, M. W. (1997). Binary image reconstruction via 2-D Viterbi search. In IEEE International Conference on Image Processing (Vol. 1, pp. 181-184). IEEE Comp Soc.

Binary image reconstruction via 2-D Viterbi search. / Miller, C.; Hunt, B. R.; Neifeld, Mark A; Marcellin, Michael W.

IEEE International Conference on Image Processing. Vol. 1 IEEE Comp Soc, 1997. p. 181-184.

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

Miller, C, Hunt, BR, Neifeld, MA & Marcellin, MW 1997, Binary image reconstruction via 2-D Viterbi search. in IEEE International Conference on Image Processing. vol. 1, IEEE Comp Soc, pp. 181-184, Proceedings of the 1997 International Conference on Image Processing. Part 2 (of 3), Santa Barbara, CA, USA, 10/26/97.
Miller C, Hunt BR, Neifeld MA, Marcellin MW. Binary image reconstruction via 2-D Viterbi search. In IEEE International Conference on Image Processing. Vol. 1. IEEE Comp Soc. 1997. p. 181-184
Miller, C. ; Hunt, B. R. ; Neifeld, Mark A ; Marcellin, Michael W. / Binary image reconstruction via 2-D Viterbi search. IEEE International Conference on Image Processing. Vol. 1 IEEE Comp Soc, 1997. pp. 181-184
@inproceedings{bd4ba13f78274f9fa44c2cb1c753c7e3,
title = "Binary image reconstruction via 2-D Viterbi search",
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 = "1997",
language = "English (US)",
volume = "1",
pages = "181--184",
booktitle = "IEEE International Conference on Image Processing",
publisher = "IEEE Comp Soc",

}

TY - GEN

T1 - Binary image reconstruction via 2-D Viterbi search

AU - Miller, C.

AU - Hunt, B. R.

AU - Neifeld, Mark A

AU - Marcellin, Michael W

PY - 1997

Y1 - 1997

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=0031341444&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031341444&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0031341444

VL - 1

SP - 181

EP - 184

BT - IEEE International Conference on Image Processing

PB - IEEE Comp Soc

ER -