Parallel image classification using multiscale Markov random fields

Z. Kato, Marc Berthod, Josiane Zerubia

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

16 Citations (Scopus)

Abstract

In this paper, we are interested in massively parallel multiscale relaxation algorithms applied to image classification. First, we present a classical multiscale model applied to supervised image classification. The model consists of a label pyramid and a whole observation field. The potential functions of the coarse grid are derived by simple computations. Then, we propose another scheme introducing a local interaction between two neighbor grids in the label pyramid. This is a way to incorporate cliques with far apart sites for a reasonable price. Finally we present the results on noisy synthetic data and on a SPOT image obtained by different relaxation methods using these models.

Original languageEnglish
Title of host publicationProceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing
PublisherPubl by IEEE
Volume5
ISBN (Print)0780309464
Publication statusPublished - 1993
EventIEEE International Conference on Acoustics, Speech and Signal Processing, Part 5 (of 5) - Minneapolis, MN, USA
Duration: Apr 27 1993Apr 30 1993

Other

OtherIEEE International Conference on Acoustics, Speech and Signal Processing, Part 5 (of 5)
CityMinneapolis, MN, USA
Period4/27/934/30/93

Fingerprint

image classification
Image classification
pyramids
grids
multiscale models
SPOT (French satellite)
Labels
interactions

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Kato, Z., Berthod, M., & Zerubia, J. (1993). Parallel image classification using multiscale Markov random fields. In Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing (Vol. 5). Publ by IEEE.

Parallel image classification using multiscale Markov random fields. / Kato, Z.; Berthod, Marc; Zerubia, Josiane.

Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. Vol. 5 Publ by IEEE, 1993.

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

Kato, Z, Berthod, M & Zerubia, J 1993, Parallel image classification using multiscale Markov random fields. in Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. vol. 5, Publ by IEEE, IEEE International Conference on Acoustics, Speech and Signal Processing, Part 5 (of 5), Minneapolis, MN, USA, 4/27/93.
Kato Z, Berthod M, Zerubia J. Parallel image classification using multiscale Markov random fields. In Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. Vol. 5. Publ by IEEE. 1993
Kato, Z. ; Berthod, Marc ; Zerubia, Josiane. / Parallel image classification using multiscale Markov random fields. Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. Vol. 5 Publ by IEEE, 1993.
@inproceedings{5de994aa01334fa2b5fba33037317ace,
title = "Parallel image classification using multiscale Markov random fields",
abstract = "In this paper, we are interested in massively parallel multiscale relaxation algorithms applied to image classification. First, we present a classical multiscale model applied to supervised image classification. The model consists of a label pyramid and a whole observation field. The potential functions of the coarse grid are derived by simple computations. Then, we propose another scheme introducing a local interaction between two neighbor grids in the label pyramid. This is a way to incorporate cliques with far apart sites for a reasonable price. Finally we present the results on noisy synthetic data and on a SPOT image obtained by different relaxation methods using these models.",
author = "Z. Kato and Marc Berthod and Josiane Zerubia",
year = "1993",
language = "English",
isbn = "0780309464",
volume = "5",
booktitle = "Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Parallel image classification using multiscale Markov random fields

AU - Kato, Z.

AU - Berthod, Marc

AU - Zerubia, Josiane

PY - 1993

Y1 - 1993

N2 - In this paper, we are interested in massively parallel multiscale relaxation algorithms applied to image classification. First, we present a classical multiscale model applied to supervised image classification. The model consists of a label pyramid and a whole observation field. The potential functions of the coarse grid are derived by simple computations. Then, we propose another scheme introducing a local interaction between two neighbor grids in the label pyramid. This is a way to incorporate cliques with far apart sites for a reasonable price. Finally we present the results on noisy synthetic data and on a SPOT image obtained by different relaxation methods using these models.

AB - In this paper, we are interested in massively parallel multiscale relaxation algorithms applied to image classification. First, we present a classical multiscale model applied to supervised image classification. The model consists of a label pyramid and a whole observation field. The potential functions of the coarse grid are derived by simple computations. Then, we propose another scheme introducing a local interaction between two neighbor grids in the label pyramid. This is a way to incorporate cliques with far apart sites for a reasonable price. Finally we present the results on noisy synthetic data and on a SPOT image obtained by different relaxation methods using these models.

UR - http://www.scopus.com/inward/record.url?scp=0027266514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027266514&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0780309464

VL - 5

BT - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

PB - Publ by IEEE

ER -