Algorithms and Architectures

  • I. Pitas
  • A. N. Venetsanopoulos
Chapter
Part of the The Springer International Series in Engineering and Computer Science book series (SECS, volume 84)

Abstract

One of the main problems in linear as well in the nonlinear image processing is the computational complexity and the execution speed of the image processing routines, especially for applications where real-time image processing is required. This problem mainly results from the amount of the data to be processed (e.g., 256K bytes for a single 512×512 8 bit image) as well as from the number of operations required per output pixel. Such operations are usually comparisons, additions, multiplications, and nonlinear function evaluations. Usually the number of operations required per output pixel is not very high. However, if this number is multiplied by the output image size (in pixels), the total amount of computations required is tremendous. Therefore, even simple nonlinear filters (e.g., median filters, erosion, dilation), which require only comparisons, are relatively slow for fast image processing applications. There are two solutions to the requirement for an increase of the speed of image processing routines. The first one is to construct fast algorithms for linear and nonlinear image processing. These algorithms can be relatively fast on general purpose computers for fast nonreal-time digital image processing. If real-time processing is a must, the only solution is to build image processors, whose architecture is optimized for image processing applications. If the performance of such image processors is not adequate, parallel image processing is the solution.

Keywords

Median Filter Finite Impulse Response Digital Filter Systolic Array Finite Impulse Response Filter 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    D.E. Knuth, The art of computer programming, vol. 3, Addison-Wesley, 1973.Google Scholar
  2. [2]
    E.Horowitz, S. Sahni, Fundamentals of Computer Algorithms, Computer Science Press, 1984.Google Scholar
  3. [3]
    T.S. Huang, G.J. Yang, G.Y. Tang, “A fast two-dimensional median filtering algorithm”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-27, no. 1, pp. 13–18, 1979.CrossRefGoogle Scholar
  4. [4]
    F. Pasian, “Sorting algorithms for filters based on ordered statistics: performance considerations”, Signal Processing, vol. 14, pp. 287–293, 1988.Google Scholar
  5. [5]
    M.O. Ahmad, D. Sundararajan, “A fast algorithm for two-dimensional median filtering”, IEEE Transactions on Circuits and Systems, vol. CAS-34, no. 11, pp. 1364–1374, Nov. 1987.CrossRefGoogle Scholar
  6. [6]
    A.V. Aho, J.E. Hoperoft, J.D. Ullman, The design and analysis of computer algorithms, Addison-Wesley, 1974.Google Scholar
  7. [7]
    P.E. Danielsson, “Getting the median faster”, Computer Graphics and Image Processing, vol. 17, pp. 71–78, 1981.CrossRefGoogle Scholar
  8. [8]
    E. Ataman, V.K. Aatre, K.M. Wong, “A fast method for real time median filtering”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-28, no. 4, pp. 415–421, Aug. 1980.CrossRefGoogle Scholar
  9. [9]
    R.T. Hoctor, S.A. Kassam, “An algorithm for order statistic determination and L-filtering”, Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, pp. 1686–1689, 1988.Google Scholar
  10. [10]
    O. Yli-Harja, J. Astola, Y. Neuvo, “Generalization of the radix-method of finding the median to weighted median, order statistic and weighted order statistic filters”, SPIE Visual Communication and Image Processing,vol.Google Scholar
  11. 1001.
    pp. 69–75, 1988.Google Scholar
  12. [11]
    V.V.B. Rao, K.S. Rao, “A new algorithm for real-time median filtering”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-34, no. 6, pp. 1674–1675, Dec. 1986.CrossRefGoogle Scholar
  13. [12]
    R.T. Hoctor, S.A. Kassam, “An algorithm and a pipelined architecture for order statistic determination and L-filtering”, IEEE Transactions on Circuits and Systems, vol. CAS-36, no. 3, pp. 344–352, March 1989.Google Scholar
  14. [13]
    I. Picas, “Fast algorithms for running ordering and max/min calculation”, IEEE Transactions on Circuits and Systems, vol CAS-36, no. 6, pp. 795804, June 1989.Google Scholar
  15. [14]
    I.Pitas, A.N. Venetsanopoulos, “A new filter structure for the implementation of certain classes of image processing operations”, IEEE Transactions on Circuits and Systems, vol. CAS-35, no. 6, pp. 636–647, June 1988.Google Scholar
  16. [15]
    J. Bee Bednar, T.L. Watt, “Alpha trimmed means and their relationship to median filters”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-32, no. 1, pp. 145–153, Feb. 1984.Google Scholar
  17. [16]
    K. (Mazer, “Design and implementation of a single-chip 1-D median filter”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-31, no. 5, pp. 1164–1168, Oct 1983.Google Scholar
  18. [17]
    J. Alsford, P. Dewer, J. Illingworth, J. Kittler, J. Lewis, K. Paler, P. Wilde, W. Thomas, “CRS image processing system with VLSI modules”, in Image processing system architectures, J. Kittler, M. Duff editors, Research Studies Press, 1985.Google Scholar
  19. [18]
    P.E. Allen, Switched capacitor circuits, Van Nostrand Reinhold, 1984.Google Scholar
  20. [19]
    D.F. Franke editor, Charge coupled devices, Springer Verlag, 1980.Google Scholar
  21. [20]
    D.R. Morgan, “Analog sorting networks ranks inputs by amplitudes and allows selection”, Electronic Design 2, Jan. 18, pp. 72–74, Jan. 1973.Google Scholar
  22. T. Jarske, P. Heinonen, Y. Neuvo, “Switched-capacitor linear median hybrid filters”, Proc. IEEE Int. Symp. on Circuits and Systems,pp. 260263, 1987.Google Scholar
  23. [22]
    J.S. Jimmy Ly, W. Harney Holmes, “Analog implementation of median filters”, IEEE Transactions on Circuits and Systems, vol. CAS-35, no. 8, pp. 1032–1033, Aug. 1988.Google Scholar
  24. [23]
    K. Chen, P. Heinonen, Q. Ye, Y. Neuvo, “Analog/Digital hybrid median filter realizations”, Proc. 1987Int. Conf. on Digital Signal Processing, pp. 349–352, Florence, Italy, 1987.Google Scholar
  25. [24]
    K. Preston, “Cellular logic computers for pattern recognition”, Computer, vol. 16, no. 1, pp. 36–47, Jan. 1983.CrossRefGoogle Scholar
  26. [25]
    M.J.B. Duff, “A cellular logic array for image processing”, Pattern Recognition, vol. 5, pp. 229–247, 1973.CrossRefGoogle Scholar
  27. [26]
    S.B. Gray, “Local properties of binary images in two dimensions”, IEEE Transactions on Computers, vol. C-20, no. 5, p. 551, May 1971.CrossRefGoogle Scholar
  28. [27]
    J.M. Herron et al., “A general purpose high speed logical transform processor”, IEEE Transactions on Computers, vol. C-31, no. 8, pp. 795–800, Aug. 1982.CrossRefGoogle Scholar
  29. [28]
    B. Kruse, “A parallel picture processing machine”, IEEE Transactions on Computers, vol. C-22, no. 12, pp. 1075–1087, Dec. 1973.CrossRefGoogle Scholar
  30. [29]
    K.E. Batcher, “Design of a massively parallel processor”, IEEE Transactions on Computers, vol. C-29, no. 9, pp. 836–840, Sept. 1980.CrossRefGoogle Scholar
  31. [30]
    S.R. Sternberg, “Biological image processing”, Computer, vol. 16, no. 1, pp. 22–34, Jan. 1983.CrossRefGoogle Scholar
  32. [31]
    J.C. Klein, J. Serra, “The texture analyzer”, Journal of Microscopy, vol. 95, pt. 2, pp. 349–356, Apr. 1972.CrossRefGoogle Scholar
  33. [32]
    S.R. Sternberg, “Pipeline architectures for image processing” in Multicomputers and image processing algorithms and programs, Academic Press, 1982.Google Scholar
  34. [33]
    S.R. Sternberg “Machine vision non-contact gaging”, Proc. Applied Machine Vision Conference, Chicago, 1984.Google Scholar
  35. [34]
    B. Picinbono, “Quadratic filters”, Proc. IEEE Int. Conf ASSP, Paris, France, pp. 298–301, 1982.Google Scholar
  36. [35]
    E. Biglieri, “Theory of Volterra processors and some applications” Proc. IEEE Int. Conf. ASSPParis, France, pp. 294–297, 1982.Google Scholar
  37. [36]
    G.L. Sicuranza, “Nonlinear digital filter realization by distributed arithmetic”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-33, no. 4, pp. 939–945, Aug. 1985.Google Scholar
  38. [37]
    G.L. Sicuranza, G. Ramponi, “Adaptive nonlinear digital filters using distributed arithmetic”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-34, no. 3, pp. 518–526, June 1986.CrossRefGoogle Scholar
  39. [38]
    H.H. Chiang, C.L. Nikias, A.N. Venetsanopoulos, “Efficient implementations of quadratic digital filters”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-34, no. 6, pp. 1511–1528, Dec. 1986.CrossRefGoogle Scholar
  40. [39]
    B.G. Mertzios, G.L. Sicuranza, A.N. Venetsanopoulos, “Efficient structures of two-dimensional quadratic digital filters”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-37, no. 5, pp. 765768, May 1989.Google Scholar
  41. [40]
    A.N. Venetsanopoulos, B.G. Mertzios, “A decomposition theorem and its implications to the design and realization of two-dimensional filters”, IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-33, no. 6, pp. 1562–1574, Dec. 1985.Google Scholar
  42. [41]
    B.G. Mertzios, A.N. Venetsanopoulos, “Modular realization of m-dimensional filters”, Signal Processing, 7, pp. 351–369, 1984.CrossRefGoogle Scholar
  43. [42]
    H.T. Kung, “Why systolic architectures”, Computers, 15, pp. 37–46, 1982.CrossRefGoogle Scholar
  44. [43]
    S.Y. Kung, “On supercomputing with systolic/wavefront array processors”, Proc. IEEE, 72, pp. 867–884, 1984.CrossRefGoogle Scholar
  45. [44]
    S.Y. Kung, VLSI array processors, Prentice Hall, 1987.Google Scholar
  46. [45]
    K. Hwang, F.A. Briggs, Computer architectures and parallel processing, McGraw-Hill, New York, 1984.Google Scholar
  47. [46]
    H.H. Lu, E.A. Lee, D.G. Messerschmitt, “Fast recursive filtering with multiple slow processing elements”, IEEE Transactions on Circuits and Systems, vol. CAS-32, no. 11, pp. 1119–1129, Nov. 1985.Google Scholar
  48. [47]
    B.G. Mertzios, A.N. Venetsanopoulos, “Implementation of quadratic digital filters via VLSI array processors”, Archiv fur Elektronik und Ubertrangungstechnik (Electronics and Communication), vol. 43, no. 3, pp. 153–157, 1989.Google Scholar
  49. [48]
    D.F. Elliot, R. Rao, Transforms, algorithms, analyses and applications, Academic Press, 1982.Google Scholar
  50. [49]
    B.G. Mertzios, V.L. Syrmos, “Implementation of digital filters via VLSI array processors”, IEE Proc., Part G, 135, pp. 78–82, 1988.MathSciNetGoogle Scholar
  51. [50]
    K.K. Parhi, D.G. Messerschmitt, “Concurrent cellular VLSI adaptive filter architectures”, IEEE Transactions on Circuits and Systems, vol. CAS-34, no. 10, pp. 1141–1151, Oct. 1987.CrossRefGoogle Scholar
  52. [51]
    V.G. Mertzios, G.L. Sicuranza, A.N. Venetsanopoulos, “Efficient structures for two-dimensional quadratic filters”, Photogrammetria, vol. 43, no. 3/4, pp. 157–166, 1989.CrossRefGoogle Scholar
  53. [52]
    A.N. Venetsanopoulos, B.G. Mertzios, “A decomposition theorem and its implications to the design and realization of two-dimensional filters”, IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-33, no. 6, pp. 1562–1574, Dec. 1985.Google Scholar
  54. [53]
    C.L. Nikias, A.P. Chrysalis, A.N. Venetsanopoulos, “The LU decomposition and its implications to the realization of two dimensional digital filters, IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-33, no. 3, pp. 694–711, June 1985.CrossRefGoogle Scholar
  55. [54]
    F.R. Gantmacher, The theory of matrices, vol. 1, Chelsea, New York, 1974.Google Scholar
  56. [55]
    H.H. Chiang, C.L. Nikias, A.N. Venetsanopoulos, “Efficient implementation of quadratic digital filters” IEEE Trans. Acoust. Speech Signal Processingvol. ASSP-34, no. 6, pp. 1511–1528, Dec. 1986.Google Scholar
  57. [56]
    Y. Lou, C.L. Nikias, A.N. Venetsanopoulos, “Efficient VLSI array processing structures for adaptive quadratic digital filters”, Circuits, Systems and Signal Processing, vol. 7, no. 2, pp. 253–273, Feb. 1988.MATHCrossRefGoogle Scholar
  58. [57]
    D. Hatzinakos, C.L. Nikias, A.N. Venetsanopoulos, “Massively parallel architecture for quadratic digital filters” IEEE Trans. on Circuits and Systemsin press.Google Scholar

Copyright information

© Springer Science+Business Media New York 1990

Authors and Affiliations

  • I. Pitas
    • 1
  • A. N. Venetsanopoulos
    • 2
  1. 1.Aristotelian University of ThessalonikiGreece
  2. 2.University of TorontoCanada

Personalised recommendations