Biomedical Parallel Image Processing on Propal 2

  • P. Roux
  • J. Richalet
Conference paper
Part of the Lecture Notes in Medical Informatics book series (LNMED, volume 17)


PROPAL 2 is a mini parallel associative computer with an SIMD architecture. Each Processing Element (PE) works in a serial bit mode, with words of variable length from 1 to 256 bits. The most sophisticated feature of this computer is a special bus called “ascenseur” which allows an easy loading of data from the host computer and performs shifting of information from one PE to others. The assembly language is an overset of the MITRA SEMS assembler. Thus programming is done in FORTRAN through assembly language subroutines in order to improve the already existing software. The ability to work on variable length words gives PROPAL 2 a great efficiency in picture processing. On large picture processing problems (scaling — filtering — convolution — feature extraction ...), the computation time Tp of PROPAL 2 is given by:
$${{T}_{p}}\underline{\tilde{\ }}1,4{{T}_{M}}/NPE$$
where TM is the computation time of a standard mini-computer (with floating point arithmetic unit) and NPE the number of elementary processors.

Compared to a SEMS MITRA 15–35, PROPAL 2 with “128 PE” being a standard configuration, the average gain to be expected is close to 90, but it depends on the problem. Real interactivity, which is a key problem in picture processing is then achieved.

A practical application will be shown.


Processing Element Parallel Processor Assembly Language Picture Processing Sequential Unit 
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.
    G.H. Barnes & Al., “The Illiac IV Computer”, IEEE T Computer Vol. C-17 August 1968, 746–757.Google Scholar
  2. 2.
    C. Timsit, “The PROPAL II Computer”, Proceedings of the 1978 International Conference on Parallel Processing, IEEE Catalog n° 78CH1321.9C.Google Scholar
  3. 3.
    A. Rosenfeld, Joan S. Weszka, “Picture Recognition and Scene Analysis”, Computer May 1976, V9N5.Google Scholar
  4. 4.
    R.P. Bishop, I.T. Young, “The Automated Classification of Mitotic Phase for Human Chromosome Spreads”, The Journal of Histochemistry and Cytochemistry Vol. 25, n° 7, 1977, 730–740.Google Scholar
  5. 5.
    P. Marthon, A. Bruel, G. Biquet, “Squelettisation par calcul d’une fonction discriminante sur un voisinage de 8 points”, 2è Congrès AFCET-INRIA Reconnaissance des formes et intelligence artificielle, Toulouse, 1979, 107–114.Google Scholar
  6. 6.
    J. Piper, E. Granum, D. Rutovitz, H. Ruttledge, “Automation of Chromosome Analysis”, 2è Congrès AFCET-IRIA Reconnaissance des formes et intelligences artificielle, Toulouse, 1979.Google Scholar
  7. R.J.P. Le Gd, “Image-processing automation for chromosome analysis”, In: Mutagen induced chromosome damage in man, Edinburg, July 1977, ( Evans, H.J., Lloyd, D.C.) Edinburg, University Press 1978, 322–325.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1982

Authors and Affiliations

  • P. Roux
    • 1
  • J. Richalet
    • 1
  1. 1.Adersa-GerbiosPalaiseauFrance

Personalised recommendations