An alternative approach to the performance evaluation of thinning algorithms for document processing applications

  • L. P. Cordella
  • A. Marcelli
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1072)


It is proposed that the performance of thinning algorithms be evaluated with reference to a task which is especially relevant in connection with the use of these algorithms in the application domain of document processing: decomposition of digital lines into meaningful parts. The stability of the decompositions obtained according to simple rules, within given classes of lines, is assumed as a performance index. Experimental results, obtained using the ETL1 database of handprinted characters, are presented, to demonstrate the representativeness of the considered parameter.


Thinning Performance evaluation Document processing 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    L. Lam, S.W. Lee and C.Y. Suen, “Thinning methodologies — A comprehensive survey”, IEEE Trans. on Patt. Anal and Mach. Intell., vol. PAMI-14, no.9, 1992, pp.869–887.Google Scholar
  2. [2]
    H. Blum, “A transformation for extracting new descriptors of shape”, in: Models for the Perception of Speech and Visual Form, (W. Watjen-Dunn, ed.), MIT Press, Cambridge:MA, 1967, pp.362–380.Google Scholar
  3. [3]
    V.K.Govindan and A.P. Shivaprasad, “A pattern adaptive thinning algorithm”, Pattern Recognition, vol.20, no.6, 1987, pp.623–637.CrossRefGoogle Scholar
  4. [4]
    R.M. Brown, T.H. Fay and C.L. Walker, “Handprinted recognition system”, Pattern Recognition, vol.21, no.2, 1988, pp.91–118.Google Scholar
  5. [5]
    X. Li and A. Basu, “Variable-resolution character thinning”, Pattern Recognition Letters, vol.12, 1991, pp.241–248.Google Scholar
  6. [6]
    A. Sirjani and G.R. Cross, “On representation of a shape's skeleton”, Pattern Recognition Letters, vol.12, 1991, pp.149–154.Google Scholar
  7. [7]
    S.W. Lu and H. Xu, “False stroke detection and elimination for character recognition”, Pattern Recognition Letters, vol.13, 1992, pp.745–755.Google Scholar
  8. [8]
    G. Boccignone, A. Chianese, L.P. Cordella and A. Marcelli, “Using skeletons for OCR”, in: Progress in Image Analysis and Processing, (V.Cantoni et al eds.), World Scientific Publishing Co., Singapore, 1990, pp.275–282.Google Scholar
  9. [9]
    H. Tamura, “A Comparison of Line Thinning Algorithms from a Digital Geometry Viewpoint”, Proc. 4th ICPR, Kyoto (JAPAN), 1978, pp.39–52.Google Scholar
  10. [10]
    C.J. Hilditch, “Comparison of Thinning Algorithms on a Parallel Processor”, IVC, vol.1, no.3, 1983, pp.115–132.Google Scholar
  11. [11]
    M.Y. Jaisimha, R.M.Haralick and D. Dori, “A Methodology for the Characterization of the Performance of Thinning Algorithms”, Proc. ICDAR '93 Tsukuba (JAPAN), October 20–22, 1993, pp. 282–286.Google Scholar
  12. [12]
    R. Plamondon, M: Bordeau, C: Chouinard, and C.Y. Suen, “Validation of Preprocessing Algorithms: A Methodology and its Application to the Design of a Thinning Algorithm for Handwritten Characters”, Proc. ICDAR '93, Tsukuba (JAPAN), October 20–22, 1993, pp. 287–290.Google Scholar
  13. [13]
    G. Boccignone, A. Chianese, L.P. Cordella and A. Marcelli, “Recovering Dynamic Information from Static Handwriting”, Pattern Recognition, vol. 26, no.3, 1993, pp.409–418.Google Scholar
  14. [14]
    C. Arcelli and G. Sanniti di Baja, “A Thinning Algorithm based on Prominence Detection”, Pattern Recognition, vol.13, no.3, 1981, pp.225–235.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1996

Authors and Affiliations

  • L. P. Cordella
    • 1
  • A. Marcelli
    • 1
  1. 1.Dipartimento di Informatica e SistemisticaUniversita' di Napoli “Federico II”NapoliItaly

Personalised recommendations