International Journal of Computer Vision

, Volume 2, Issue 3, pp 251–282 | Cite as

The image understanding architecture

  • Charles C. Weems
  • Steven P. Levitan
  • Allen R. Hanson
  • Edward M. Riseman
  • David B. Shu
  • J. Gregory Nash


This paper provides an overview of the Image Understanding Architecture (IUA), a massively parallel, multilevel system for supporting real-time image understanding applications and research in knowledge-based computer vision. The design of the IUA is motivated by considering the architectural requirements for integrated real-time vision in terms of the type of processing element, control of processing, and communication between processing elements.

The IUA integrates parallel processors operating simultaneously at three levels of computational granularity in a tightly coupled architecture. Each level of the IUA is a parallel processor that is distinctly different from the other two levels, designed to best meet the processing needs at each of the corresponding levels of abstraction in the interpretation process. Communication between levels takes place via parallel data and control paths. The processing elements within each level can also communicate with each other in parallel, via a different mechanism at each level that is designed to meet the specific communication needs of each level of abstraction.

An associative processing paradigm has been utilized as the principle control mechanism at the low and intermediate levels. It provides a simple yet general means of managing massive parallelism, through rapid responses to queries involving partial matches of processor memory to broadcast values. This has been enhanced with hardware operations that provide for global broadcast, local compare, Some/None response, responder count, and single responder select. To demonstrate how the IUA may be used for vision processing, several sample algorithms and a typical interpretation scenario on the IUA are presented.

We believe that the IUA represents a major step toward the development of a proper combination of integrated processing power, communication, and control required for real-time computer vision. A proof-of-concept prototype of 1/64th of the IUA is currently being constructed by the University of Massachusetts and Hughes Research Laboratories.


Processing Element Parallel Processor Multilevel System Massive Parallelism Single Responder 
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. M. Annarontone, E. Arnold, T. Gross, H.T. Kung, M. Lam, O. Menzilcioglu, and J.A. Webb, “The warp computer: architecture, implementation, and performance,” IEEE Trans. Comput., C-36(12), December 1987.Google Scholar
  2. D.K. Arvind, I.N. Robinson, and I.N. Parker, “A VLSI chip for real-time image processing,” Proc. IEEE Int. Symp. Circuits Syst., pp. 405–408, 1983.Google Scholar
  3. K.E. Batcher, “Design of a massively parallel processor,” IEEE Trans. Comp., C-29(9), September 1980.Google Scholar
  4. J.R. Beveridge, J. Griffith, R.R. Kohler, A.R. Hanson, and E.M. Riseman, “Segmenting images using localized histograms and region merging,” Internat. J. of Computer Vision 2(3), January 1989.Google Scholar
  5. J.B.Burns, A.R.Hanson, and E.M.Riseman, “Extracting straight lines,” IEEE Trans. PAMI 8:425–455, 1986.Google Scholar
  6. W.Crowther, J.Goodhue, E.Starr, R.Thomas, W.Milliken, and T.Blackadar, “Performance measurements on a 128-node butterfly parallel processor,” Proc. Int. Conf. Parallel Processing, IEEE Computer Soc. Press: Washington, DC, pp. 531–540, 1985.Google Scholar
  7. R.Davis, and D.Thomas, “Geometric arithmetic parallel processor-systolic array chip meets the demands of heavyduty processing,” Electronic Design, October 31, pp. 207–218, 1984.Google Scholar
  8. B.A.Draper, R.T.Collins, J.Brolio, J.Griffith, A.R.Hanson, and E.M.Riseman, “Tools and experiments in the knowledge-directed interpretation of road scenes,” Proc. Image Uniderstanding Workshop, Morgan Kaufmann: Los Altos, CA, 1987.Google Scholar
  9. B.A. Draper, R.T. Collins, J. Brolio, J. Griffith, A.R. Hanson, and E.M. Riseman, “The schema system,” Internat. J. of Computer Vision 2(3), January 1989.Google Scholar
  10. M.J.B. Duff, “Review of the CLIP image proceeding system,” Proc. Nat. Comput. Conf., AFIPS, pp. 1055–1060, 1978.Google Scholar
  11. M.J.B.Duff (ed.), Intermediate-Level Image Processing. Academic Press: London, 1986.Google Scholar
  12. Encore Computer Corp., promotional literature, Marlborough, MA, 1986.Google Scholar
  13. ErmanL., et al., “The Hearsay-II Speech-Understanding System: integrating knowledge to resolve uncertainty,” Computing Surveys 12:213–253, 1980.Google Scholar
  14. C.C.Foster, Content Addressable Parallel Processors. New York: Van Nostrand Reinhold, 1976.Google Scholar
  15. A.R. Hanson, and E.M. Riseman, “Preprocessing cones: a computational structure for scene analysis,” COINS Tech. Rept. 74C-7, Univ. of Massachusetts at Amherst, September 1974.Google Scholar
  16. A.R.Hanson, and E.M.Riseman, “VISIONS: A computer system for interpreting scenes.” In Computer Vision Systems, A.R.Hanson and E.M.Riseman (eds.), New York: Academic Press, pp. 303–333, 1978.Google Scholar
  17. A.R.Hanson, and E.M.Riseman, “Segmentation of natural scenes.” In Computer Vision Systems, A.R.Hanson, and E.M.Riseman (eds.), New York: Academic Press, pp. 129–163, 1978.Google Scholar
  18. A.R.Hanson, and E.M.Riseman, “Processing cones: a computational structure for scene analysis for image analysis.” In Structured Computer Vision, S.Tanimoto and A.Klinger (eds.), New York: Academic Press, 1980.Google Scholar
  19. A.R.Hanson, and E.M.Riseman, “A methodology for the development of general knowledge-based vision systems.” In Vision, Brain, and Cooperative Computation, M.Arbib and A.Hanson (eds.), MIT Press: Cambridge, 1986.Google Scholar
  20. A.R. Hanson, and E.M. Riseman, “From image measurements to object hypotheses,” COINS Tech. Rept. 87-129, Univ. Massachusetts at Amherst, December 1987.Google Scholar
  21. D.W.Hillis, The Connection Machine. MIT Press: Cambridge, 1986.Google Scholar
  22. D.J.Hunt, “The ICL DAP and its application to image processing.” In Languages and Architectures for Image Processors, M.J.B.Duff and S.Levialdi (eds.), Academic Press: London, 1981.Google Scholar
  23. V.K.P. Kumar and C.S. Raghavendra, “Array processor with multiple broadcasting,” Proc. 12th Ann. Int. Symp. Comput. Architecture, June, 1985.Google Scholar
  24. N.B. Lehrer, G. Reynolds, and J. Griffith, “A method for initial hypothesis formation in image understanding,” Proc. Int. Conf. Computer Vision, London, pp. 578–585, June 1987. Computer Society Press.Google Scholar
  25. S.P. Levitan, “Parallel algorithms and architectures: a programmers perspective.” PhD dissertation, Computer and Information Science Department; also, COINS Tech. Rept. 84-11, University of Massachusetts at Amherst, May 1984.Google Scholar
  26. S.P.Levitan, “Measuring communication structures in parallel architectures and algorithms.” In The Characteristics of Parallel Algorithms L.Jamieson, D.Gannon, and R.Douglass (eds.), MIT Press: Cambridge, 1987.Google Scholar
  27. H.Li and M.Moresca, “Polymorphic Torus Network,” Proc. Intl. Conf. Parallel Processing, Pennsylvania State University Press: State College, Pennsylvania, 1987.Google Scholar
  28. R. Miller, V.K.P. Kumar, Dionisios Reisis, and Q.F. Stout, USC Tech. Rept. IRIS #229, Univ. of Southern California, Los Angeles, CA, March 1987.Google Scholar
  29. P.A.Nagin, A.R.Hanson, and E.M.Riseman, “Studies in global and local histogram-guided relaxation algorithms,” IEEE Trans. PAMI 4:263–277, 1982.Google Scholar
  30. NCube Corp, Promotional literature, Beaverton, OR, 1985.Google Scholar
  31. H.P.Nii, “The blackboard model of problem solving and the evolution of blackboard architectures,” AI Magazine, vol 7, no. 2, pp. 38–53, 1986.Google Scholar
  32. C.C. Parma, A.R. Hanson, and E.M. Riseman, “Experiments in schema-driven interpretation of a natural scene,” COINS Tech. Rept. 80-10, Univ. Massachusetts at Amherst, April 1980. Also in: NATO Advanced Study Institute on Digital Image Processing, R. Haralick and J.C. Simon (eds.), Bonas, France, 1980.Google Scholar
  33. G.Pfister, W.Brantley, D.George, S.Harvey, W.Kleinfelder, K.McAuliffe, E.Melton, V.Norton, and J.Weiss, “The IBM Research Parallel Processor Prototype (RP3): introduction and architecture,” Proc. Intl. Conf. Parallel Processing, IEEE Computer Soc. Press: Washington, DC, pp. 764–771, 1985.Google Scholar
  34. The Random House Unabridged Dictionary of the English Language 2nd ed., S.B.Flexner, and L.C.Hauck (eds.), Random House: New York, 1987.Google Scholar
  35. J. Rattner, “Concurrent processing: a new direction in scientific computing,” Proc. Natl. Comput. Conf., 1985.Google Scholar
  36. G. Reynolds, N. Irwin, A.R. Hanson, and E.M. Riseman, “Hierarchical knowledge-directed object extraction using a combined region and line representation,” Proc. Workshop Computer Vision: Representation and Control, Annapolis, Maryland, pp. 238–247, April 30–May 2, 1984.Google Scholar
  37. E.M. Riseman and A.R. Hanson, “Computer vision research at the University of Massachusetts-themes and figures,” Internat. J. of Computer Vision 2(3), January 1989.Google Scholar
  38. A. Rosenfeld, “The prism machine: an alternative to the pryamid,” J. Parallel and Distributed Comput. 3: 404–411.Google Scholar
  39. A.Rosenfeld, “A report on the DARPA Image Understanding Architectures Workshop,” Proc. DARPA Image Understanding Workshop, Morgan Kaufmann: Los Altos, CA, 1987.Google Scholar
  40. L.Seitz, “The cosmic cube,” Communications of the ACM, 28–1, pp. 22–33, January 1985.Google Scholar
  41. Sequent Computer Systems, promotional literature, Beaverton, OR, 1986.Google Scholar
  42. S. Tanimoto, “A pyramidal approach to parallel processing,” Proc. 10th Ann. Int. Symp. Comput. Architecture, Stockholm, Sweden, June 1983.Google Scholar
  43. L.Uhr, “Layered recognition cone networks that preprocess, classify and describe,” IEEE Trans. Comput., 21:758–768, 1972.Google Scholar
  44. L.M.Uhr, Parallel Computer Vision, Academic Press: Orlando, FL, 1987.Google Scholar
  45. C.C. Weems, “Image processing on a content addressable array parallel processor,” PhD dissertation, Computer and Information Science Department; also, COINS Tech. Rept. 84-14, Univ. of Massachusetts at Amherst, September 1984.Google Scholar
  46. C.C. Weems, S.P. Levitan, C.C. Foster, E.M. Riseman, D.T. Lawton, and A.R. Hanson, “Development and construction of a Content Addressable Array Parallel Processor (CAAPP) for knowledge-based image interpretation,” Proc. Workshop on Algorithm-Guided Parallel Architectures for Automatic Target Recognition, Leesburg, VA, July 16–18, pp. 329–359, 1984.Google Scholar
  47. C.C. Weems, “The Content Addressable Array Parallel Processor: architectural evaluation and enhancement,” Proc. IEEE Int. Conf. Comput. Design: VLSI in Computers, Port Chester, New York, October 7–10, pp. 500–503, 1985.Google Scholar
  48. C.C. Weems, A.R. Hanson, E.M. Riseman, and A. Rosenfeld. “An integrated image understanding benchmark: recognition of a \(2{\raise0.5ex\hbox{$\scriptstyle 1$}\kern-0.1em/\kern-0.15em\lower0.25ex\hbox{$\scriptstyle 2$}}\)D ‘Mobile’,” Proc. IEEE Conf. Comput. Vision and Pattern Recognition, Ann Arbor, MI, June 5–9, 1988.Google Scholar
  49. T.E. Weymouth, “Using object descriptions in a schema network for machine vision.” PhD dissertation, Computer and Information Science Department; also, COINS Tech. Rept. 86-24, Univ. of Massachusetts at Amherst, 1986.Google Scholar

Copyright information

© Kluwer Academic Publishers 1989

Authors and Affiliations

  • Charles C. Weems
    • 1
  • Steven P. Levitan
    • 1
  • Allen R. Hanson
    • 1
  • Edward M. Riseman
    • 1
  • David B. Shu
    • 2
  • J. Gregory Nash
    • 2
  1. 1.Department of Computer and Information ScienceUniversity of MassachusettsAmherstUSA
  2. 2.Hughes Research LaboratoriesMalibuUSA

Personalised recommendations