SMAE: A Multiprocessor System for Architecture Emulation in Intermediate Level Image Processing Tasks

  • A. Machì
Part of the International Centre for Mechanical Sciences book series (CISM, volume 307)


A wide class of image analysis tasks are devoted to extract from data expressed in iconic form (images) descriptors expressed in not iconic form (graphs, lists of attributes). As well known, such tasks require both local and global operations over the input image and are not always efficiently managed neither by massive SIMD processors, nor by various types of MIMD processors so that a number of novel computer architectures are presently being experimented.

A research project has been started at IFCAI to build a multiprocessor system able to emulate and evaluate on a common testbed different strategies of coupling architecture facilities, algorithms and languages presently used in mid-level image processing.

The system is called SMAE (SIMD / MIMD Architecture Emulator) and it is based on a PRAM (Parallel Random Access Memory) parellel architecture model implemented on multiple bus structure and augmented with a Near_Neighbour linear network.

In the paper some architectural issues on parallel image processing are discussed, a description of the machine hardware and of some its emulation possibilities follows.


Input Image Adjacency Matrix Shared Memory Multiprocessor System Memory Bank 
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.
    Tanimoto, S.: Architectural solutions for Intermediate Level Vision” in: Real Time Object and Environment Measurement and Recognition, (ed.Jain A. K.), Springer Verlag NATO ASI Scries F 1988.Google Scholar
  2. 2.
    Cantoni, V., Carrioli, L., et al.: “Object Recognition in multiresolution systems” in: Real Time Object and Environment Measurement and Recognition, (ed. Jain A. K.), Springer Verlag NATO ASI Series F 1988.Google Scholar
  3. 3.
    Cantoni, V, Guerra C and Lcvialdi, S.: “Towards an Evaluation of an Image Processing System”, in: Computing Structures for Image Processing, (ed. Duff, M. J. B.), Academic Press London 1983.Google Scholar
  4. 4.
    Fountain, T. J.: “Introducing Local Autonomy to Processor Array”. Proc. CAIP Workshop on Machine Vision, (ed. Freeman, H.), Academic Press New York 1987.Google Scholar
  5. 5.
    Intermediate Level Image Processing (ed. Duff, M, J., B.), Academic Press London,1986.Google Scholar
  6. 6.
    Machí, A. , Di Gesù, V. , Lupo, F.: “Performance of the mesh architecture in the analysis of sparse images” in: Data Analysis in Astronomy III, (cd Di Gesù V. et al) Plenum Press New York, in press.Google Scholar
  7. 7.
    Duff, M.,J.,B., “How Not To Evaluate Image Processors” in: Evaluation of Multicomputers for Image Processing, (ed. Uhr, L., Preston, K., Lcvialdi, S., Duff, M.,J.,B.,) Academic Press London, 1985.Google Scholar

Copyright information

© Springer-Verlag Wien 1989

Authors and Affiliations

  • A. Machì
    • 1
  1. 1.Istituto di Fisica Cosmica ed Applicata dell’InformaticaCNRPalermoItaly

Personalised recommendations