Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

Architectures massivement paralleles pour la vision artificielle

Massively parallel architecture for the artificial vision

  • 20 Accesses

Résumé

Cet article présente un certain nombre de machines massivement parallèles plus particulièrement destinées aux applications de traitement ďimages et de vision artificielle. Après une description de principes de traitements reposant sur divers concepts architecturaux (traitement pipeline, parallélisme des données), trois machines représentatives de classes architecturales sont particulièrement ètudiées : MPP (tableau de processeurs), la Connection Machine (hypercube) et Sphinx (pyramide). Une étude comparative de ľ implantation ďun probléme typique ďanalyse ďimages est réalisée. Ľarticle discute également de ľapport des techniques ď intégration de circuit à la réalisation de telles machines.

Abstract

This paper presents some massively parallel computers, especially those dedicated to image processing and artificial vision. After a discussion on the information processing principle in an architectural point of view, we analyse three representative machines: MPP (processor array), Connection Machine (hypercube machine) and Sphinx (pyramid machine). A comparative study of the implementation of image analysis is carried out. This paper also emphasizes the influence of integration technologies to the realization of the massively parallel computers.

This is a preview of subscription content, log in to check access.

Bibliography

  1. [Blelloch, 89]

    Blelloch (G.). Scans as primitive parallel operations. Proceeding of the 1987 International Conference on Parallel Processing,IEEE Trans.on Computers (1989),38, n° 11, pp. 1526–1538.

  2. [Burt, 81]

    Burt (P.), Hong (T.H.), Rosenfeld (A.). Segmentation and estimation of image region properties through cooperative hierarchical computation.IEEE Trans.on Systems, Man and Cybernetics (Dec 82),15, n° 12, pp. 802–825.

  3. [Cantoni, 85]

    Cantonii et al. (V.). The papia image analysis system. 2nd Int. Symp. on Optical and Electrooptical Applied Science and Engeneering. Cannes, France (Dec. 85).

  4. [Cantoni, 86]

    Cantoni (V.),Levialdi (S.) Eds. Pyramidal systems for computer vision, Springer Verlag,NATO ASI Series (1986).

  5. [Clermont, 87a]

    Clermont (Ph.),Mérigot (A.). Real time synchronization.In a multi-SIMD massively parallel machine.CAPAMI conf. Seattle, WA (oct. 87).

  6. [Clermont, 87b]

    Clermont (Ph.),Belaid (A.),Mérigot (A.). A parallel pyramidal algorithm to determine curve orientation in Image analysis and processing II. [Cantoni, V. Di Gesu, S. Levialdi Eds.]Plenum (1988), pp. 353–360.

  7. [Clermont, 90]

    Clermont (Ph.),Méhat (A.),Roussel (J.C.),Zavidovique (B.). Parallel implementation of a region merging algorithm on a pyramid machine.In Problems in image analysis and processing (Cantoni V., eds),World scientific (1990), pp. 721–728.

  8. [Clermont, 89b]

    Clermont (Ph.). Méhodes de programmation de machines cellulaires pyramidales.These de Doctoral, Univ. Paris VI (1989).

  9. [Duff, 76]

    Duff (M.L.B.). CLIP 4, a large scale integrated circuit array parallel processor.Proc. 3rd Int. Conf. on Pattern Recognition (1976), pp. 728–733.

  10. [Flanders, 78]

    Flanders et al. Efficient high speed computing with the distributed array processor.In High speed computer and algorithm organization, [Kuck and al. Eds],Academic Press (1978).

  11. [Hillis, 85]

    Hillis (D.). The connection machine.MIT Press (1985).

  12. [Little, 89]

    Little (J.),Blelloch (G.E.),Cass (T.A.). Algorithmic techniques for computer vision on a fine-grained parallel machine.IEEE Trans. on Pattern Analysis and Machine Intelligence (mars 1989).

  13. [Méhat, 88]

    Méhat (J.), Mérigot (A.). Controling and programming theSphinx multi-MIMD pyramid machine.Proc. 2nd Symp. on Frontiers of Massively Parallel Computation, Fairfax, Virginia (oct. 1988), pp. 423–428.

  14. [Méhat, 89]

    Méhat (J.). Contrôle et programmation parallèle des architectures pyramidales.Thése de Doctorat, Univ. Paris VIII (1989).

  15. [Méhat, 85]

    Mérigot (A.),Zavidovique (B.),Devos (R).Sphinx, a pyramidal approach to parallel image processing.IEEE Workshop on CADAIDM, Miami, 18–20 Nov. (1985), pp. 107–111.

  16. [Méhat, 86a]

    Mérigot (A.),Clermont (P.),Méhat (J.),Devos (R),Zavidovique (B.). A pyramidal system for image processing.In Pyramidal systems for image processing and computer vision

  17. [Cantoni, S. Levialdi, Eds.]

    , Springer Verlag (1986), pp. 109–124.

  18. [Méhat, 86b]

    Mérigot (A.),Garda (P.),Zavidovique (B.),Devos (R). Interlayer communication in mimd pyramidal computer.8th Int. Conf. on Pattern Recognition, Paris, 28–31 octobre 1986.

  19. [Mérigot, 88a]

    Mérigot (A.). Designing memories for cellular processors. In Multi-Computer Vision [S. Levialdi Ed.],Academic Press (1988), pp. 171–183.

  20. [Mérigot, 88b]

    Mérigot (A.). map, Manuel de référence, Rapport Interne, ief,Univ. Paris Sud (1988).

  21. [Ni, 89]

    Ni (Y.),Méhat (A.),Devos (R). Designing theSphinx vlsi processing element chip.5th Int. Conf. on Image Analysis and Processing, Positano, Italie, septembre 1989.

  22. [Ni, 90]

    Ni (Y.). Contribution à ľétude des architectures massivement parallèles pyramidales : Réalisation ďun circuit vlsi multiprocesseur et définition ďun modéle de contrôle dynamique, Thèse de doctorat de ľUniv. Paris Sud (1990).

  23. [Potter, 85]

    Potter (J.L.). The massively parallel processor,MIT Press (1985).

  24. [Miller, 84]

    Miller (R.),Stout (R). Convexity algorithms for pyramid computers.Proc. of the International Conf. on Parallel Processing, 1984.

  25. [Miller, 87]

    Miller (R.),Stout (Q.R). Data movement techniques for the pyramid computer. SIAM J. Computing (FéV. 1987), 16, n° 1.

  26. [Preston, 84]

    Preston (K.),Duff (M.). Modern cellular automata.Plenum (1984).

  27. [Reeves, 84]

    Reeves (A.P.). Parallel Pascal : an extended Pascal for parallel computer.Journal of parallel and distributed computing (1984),1.

  28. [Rosenfeld, 84]

    Rosenfeld (A.). Multiresolution image processing and analysis.Springer Verlag (1984).

  29. [Schaefer, 87]

    Scheafer (D.H.),Ho (P.),Boyd (J.),Vallejos (C). The gam pyramid.In Parallel computer vision [L. Uhr Ed.],Academic Press (1987), pp. 15–42.

  30. [Tanimoto, 83]

    Tanimoto (S.). A pyramidal approach to parallel processing.Proc. 10th Int. Symp. on Computer Architecture, Stockholm (1983), pp. 253–256.

  31. [Tucker, 88]

    Tucker (L.W.),Robertson (G.G.). Architecture and Applications of the connection machine.IEEE Computer (août 1988), pp. 26–38.

  32. [Uhr, 87]

    Uhr (L.). Highly parallel, hierarchical, recognition cone perceptual structures.In Parallel computer vision [L. Uhr Ed.],Academic Press (1987), pp. 249–292.

Download references

Author information

Correspondence to Alain MéRIGOT or Yang Ni or Francis Devos.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

MéRIGOT, A., Ni, Y. & Devos, F. Architectures massivement paralleles pour la vision artificielle. Ann. Télécommun. 46, 78–89 (1991). https://doi.org/10.1007/BF02995438

Download citation

Mots clés

  • Système parallèle
  • Vision ordinateur
  • Traitement image
  • Architecture ordinateur

Key words

  • Parallel system
  • Computer vision
  • Image processing
  • Computer architecture