A RISC-type structural design of the Hermes multiprocessor kernel

  • Dimitris Fotakis
  • Nikolaos Bourbakis
Session 10: Algorithms, Architectures And Performance III
Part of the Lecture Notes in Computer Science book series (LNCS, volume 297)


HERMES is a heterogeneous, real-time, reconfigurable, multiprocessor vision machine, consisted of [N2/4] microprocessor-nodes in a 2-D array structure. It receives the image data directly from the environment, using a 2-D photoarray of NxN cells, and processes them in a parallel-hierarchical (bottom-up and top-down) manner without using a host computer for its function.

This paper deals with the RISC-type structural organization of the HERMES multi-processor kernel and with some ideas for its VLSI geometric configuration. The kernel of the HERMES vision system consists of four Z-adjacent microprocessornodes which occupy a square type area.


Control Line Slave Node VLSI Design Picture Information Slave Processor 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    S. Tanimoto and A. Klinger,"Structured Computer Vision", Academic Press, N.Y., 1980.Google Scholar
  2. [2]
    N.G.Bourbakis and D.K.Fotakis,"HERMES — A hierarchical, multiprocessor machine vision system architecture", submittedGoogle Scholar
  3. [3]
    N.G.Bourbakis,"Design of real-time supercomputing vision architectures", Proc. of IEEE Conf. on SUPERCOMPUTING, May 1987, Santa Clara, CA.Google Scholar
  4. [4]
    S.Tanimoto,"A pyramidal approach to parallel processing", Proc. Int. Symp. on Comp. Archit.,Stockholm,Sweden,June 1983, pp. 372–378.Google Scholar
  5. [5]
    M. Duff,"Computing Structures for Image Processing", Academic Press, N.Y. 1983.Google Scholar
  6. [6]
    E.P.Danielson and S.Levialdi,"Computers architectures for pictorial information systems",IEEE Computer, Nov.1981,pp.53–67.Google Scholar
  7. [7]
    G.H.Granland,"GOP image Processor",Picture Processing Lab, Linkoeping University, Sweden.Google Scholar
  8. [8]
    G.J. Li and B.W. Wah,"The design of optimal systolic arrays",IEEE Trans. on Computers, Vol.34,No.1,pp.66–77.1985.Google Scholar
  9. [9]
    J.L. Hennessy,"VLSI Processor architecture",IEEE Trans. on Computers,vol33,No.12,1221–1246,1984.Google Scholar
  10. [10]
    N. Dimopoulos,"On the structure of the HOMOGENEOUS multiprocessor, IEEE Trans. on Computers,vol.34,pp.141–150,1985.Google Scholar
  11. [11]
    N.G.Bourbakis and H.Nguyen,"TALOS — A distributed image analysis/synthesis multiprocessor system", submitted to IEEE Conf.,1987.Google Scholar
  12. [12]
    N.G.Bourbakis and P.Ligomenides,"A real-time, hierarchical multimicroprocessor vision system",Proc.IEEE Conf.on CVPR 86,June 1986,Miami,Florida, pp.381–387.Google Scholar
  13. [13]
    G.Fritsch et. al.,"EMS-85-The Erlangen Multiprocessor system for board spectrum of applications",Proc.IEEE Conf. on Supercomputers, 1985.Google Scholar
  14. [14]
    L.M.Ni and A.K.Jain,"A VLSI systolic architecture for pattern clustering", IEEE Trans. on PAMI,vol.7,No.1,1986.Google Scholar
  15. [15]
    C.L.Seitz,"Concurrent VLSI architectures",IEEE Trans. on Computers,vol.33,No.12,1984.Google Scholar
  16. [16]
    N.G.Bourbakis and D.K.Fotakis,"Structure and operation of the HERMES Multiprocessor kernel",Proc. IEEE Conf. on NCC, June 1987,Chicago,IL.Google Scholar
  17. [17]
    N.G. Bourbakis and C. Vaitsos,"A multimicroprocessor tree network configuration used on robot vision systems",Digital Techniques,S. Tzafestas ed. Elsevier Sc.Pub., North Holland,IMACS 1985,pp.483–490.Google Scholar
  18. [18]
    N.G.Bourbakis and P.Ligomenides,"High performance architectures for realtime multilevel picture information systems",Proc. IEEE Conf. on LFA, June 1985,Mallorca, Spain,pp.271–276.Google Scholar
  19. [19]
    D. Panagiotopoulos and N. Bourbakis,"The VLSI design of a 2-D image processing array",Int.Journal on Microprogramming and Microprocessing,vol.14,No.3–4,pp.125–132,1984.Google Scholar
  20. [20]
    "Clipper 33 MHz performance",FAIRCHILD CO., CA.,1987.Google Scholar
  21. [21]
    D.K. DuBose,D.K. Fotakis and D. Tabak,"A microcoded RISC", Computer Architecture News,vol.14,No.3,pp.5–16,1986, (Also,19th Microprogramming Workshop N.Y.Oct. 1986).Google Scholar
  22. [22]
    D.Tabak,"Reduced Instruction Set Computer RISC-Architecture", Research Studies Press,1987.Google Scholar
  23. [23]
    D.A. Patterson and C.H. Sequin,"A VLSI RISC",IEEE Computer,vol.15,No.9, 1982, pp.8–21.Google Scholar
  24. [24]
    "AMD 2900 Family 1985 Databook", Bipolar microprocessor logic and interface.Google Scholar
  25. [25]
    A.Klinger,"Regular decomposition and picture structure", Proc.IEEE Conf. on SMC, Dallas,Texas,pp.307–310.Google Scholar
  26. [26]
    N.G.Bourbakis and D.K.Fotakis,"KALYPSO-A picture recognition and knowledge creation language", GMU-ECE-TR-1987, to be submitted.Google Scholar
  27. [27]
    H.Freeman,"On the encoding of arbitrary geometric configuration", IRE Trans. Eletron Computing, EC-10,260–268,1961.Google Scholar
  28. [28]
    N.G.Bourbakis and D.K.Fotakis,"A heuristic scheme for recognition of progressive digital straight lines with unevenness",Proc. IEEE Workshop on LFA, Aug.1988.Google Scholar
  29. [29]
    N.G.Bourbakis,"MYLOS-A multilevel operating system for the HERMES machine", to be submitted.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • Dimitris Fotakis
    • 1
    • 2
  • Nikolaos Bourbakis
    • 1
    • 2
  1. 1.Dept. of Electrical and Computer EngineeringGeorge Mason UniversityFairfax
  2. 2.Computer Technology InstitutePatrasGreece

Personalised recommendations