Parallel algorithms for labeling image components

  • Wen-Jing Hsu
  • Xiaola Lin
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 450)


We study the problem of labeling connected components of an N1/2×N1/2 image using parallel computers. The following new results are presented:
  1. 1).

    Based on the CRCW PRAM (Concurrent-Read Concurrent-Write Parallel Random Access Machine) model, we show that Θ(N/logN) processors are necessary and sufficient to solve the image component problem in O(logN) time. The best known algorithm uses N processors to achieve the same time bound.

  2. 2).

    We show that Θ(N/log2N) hypercube or shuffle processors are necessary and sufficient to solve the same problem in O(log2N) time. This new result improves the processor requirement of the best existing algorithm by a factor of O(log2N).

  3. 3).

    We present a new mesh computer algorithm aimed to improve the utilization of the processor resource while maintaining reasonable speedup over the sequential algorithm. The algorithm that is presented uses only O(N1/2) processors and executes in O(N1/2logN) time. This time complexity is a factor of O(logN) away from the optimal time complexity of O(N1/2); however, the number of processors is reduced by a factor of O(N1/2) from O(N).

  4. 4).

    We show that these algorithms can be extended to compute components in k-dimension and to compute components based on generalized connectivities.



Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [Akl85]
    S. Akl, Parallel Sorting, Academic Press, Orlando, Fla., 1985.Google Scholar
  2. [Batcher68]
    K. E. Batcher, Sorting Networks and Their Applications, Proceedings of the AFIPS 1968 Spring Joint Computer Conference, Atlantic City, New Jersey, pp. 307–314, AFIPS Press, Montvale, N. J., 1968.Google Scholar
  3. [Cypher89]
    R. Cypher, J. L. C. Sanz, and L. Snyder, Hypercube and Shuffle-Exchange Algorithms for Image Component Labeling, J. of Algorithms, 10, pp. 140–150, (1989).Google Scholar
  4. [Cypher87]
    R. Cypher, J. L. C. Sanz, and L. Snyder, EREW PRAM and Mesh Connected Computer Algorithms for Image Component Labeling, in "1987 Computer Society Workshop on Computer Architecture, Pattern Analysis and Machine Intelligence", pp. 122–130.Google Scholar
  5. [Cypher87]
    R. Cypher, J. L. C. Sanz, and L. Snyder, Practical Algorithms for Image Component Labeling on SIMD Mesh Connected Computers, in Int. Conf. on Parallel Processing, 1987, pp. 722–779.Google Scholar
  6. [Fishburn82]
    J. P. Fishburn and R. A. Finkel, Quotient Networks, IEEE TC (Transactions on Computers), Vol. C-31, No. 4, pp. 288–295 (1982).Google Scholar
  7. [Hsu89]
    W.-J. Hsu and X. Lin, Parallel Algorithms for Labeling Image Components, TR. CPS-89-14, Dept. Comput. Sci., MSU, 1989.Google Scholar
  8. [Hummel86]
    R. Hummel, Connected Component Labeling in Image Processing with MIMD Architectures, in "Intermediate-Level Image Processing", pp. 101–127, Academic Press, New York, 1986.Google Scholar
  9. [Hung87]
    Y. Hung and A. Rosenfeld, Parallel Processing of Linear Quadtrees on a Mesh-Connected Computer, TR CAR-TR-278, Center for Automation Research, University of Maryland, March, 1987.Google Scholar
  10. [Kumar86]
    V. K. P. Kumar and M. M. Eshaghian, Parallel Geometric Algorithms for Digitized Pictures on Mesh of Trees (preliminary version), in Proc., 1986 Int. Conf. on Parallel Processing, 697–699.Google Scholar
  11. [Lim86]
    W. Lim, Fast Algorithms for Labeling Connected Components in 2-D Arrays, TR. 86.22, Thinking Machine Corp., Cambridge, MA, July 1986.Google Scholar
  12. [Miller87]
    R. Miller and Q. Stout, Data Movement Techniques for the Pyramid Computer, SIAM J. Comput. 16, No. 1 (1987), 38–60.Google Scholar
  13. [Nassimi80]
    D. Nassimi and S. Sahni, Finding Connected Components and Connected Ones on a Mesh-Connected Parallel Computer, SIAM J. Comput. 9, NO. 4 (1980), 744–757.Google Scholar
  14. [Nassimi81]
    D. Nassimi and S. Sahni, Data Broadcasting in SIMD Computers, IEEE TC, C-30, No. 2 (1981), 101–107.Google Scholar
  15. [Nassimi82]
    D. Nassimi and S. Sahni, Parallel Permutation and Sorting Algorithms and a New Generalized Connection Network, J. ACM, 29, No. 3 (1982), 642–667.CrossRefGoogle Scholar
  16. [Rosenfeld83]
    A. Rosenfeld, Parallel Image Processing Using Cellular Arrays, IEEE TC C-32 (1983), 14–20.Google Scholar
  17. [Rosenfeld82]
    A Rosenfeld and A. Kak, "Digital Picture Processing", Vols. 1–2, Academic Press, New York, 1982.Google Scholar
  18. [Shiloach82]
    Y. Shiloach and U. Vishkin, An O(logN) Parallel Connectivity Algorithm, J. Algorithms 3 (1982), 57–67.CrossRefGoogle Scholar
  19. [Stone72]
    H. Stone, Parallel Processing with the Perfect Shuffle, IEEE TC, C-20, No. 2 (1972), 153–161.Google Scholar
  20. [Stout85]
    Q. F. Stout, Properties of Divide-and Conquer Algorithms for Image Processing, in "1985 IEEE Computer Workshop on Computer Architecture for Pattern Analysis and Image Data Management", pp 203–209.Google Scholar
  21. [Tanimoto86]
    S. Tanimoto, Architectural Issues for Intermediate-Level Vision, in "Intermediate-Level Image Processing", pp. 3–16. Academic Press, New York, 1986.Google Scholar
  22. [Thompson77]
    C. D. Thompson and H. T. Kung, Sorting on a Mesh-Connected Parallel Computer, Comm. ACM, 20, 4 pp. 263–271, 1977.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1990

Authors and Affiliations

  • Wen-Jing Hsu
    • 1
  • Xiaola Lin
    • 1
  1. 1.Department of Computer ScienceMichigan State UniversityEast Lansing

Personalised recommendations