Robust Detection of Defective Parts Using Pattern Matching for Online Machine Vision System

  • Namita Singh
  • Abhishek Jaju
  • Sanjeev SharmaEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1022)


Reliable detection of defective parts is an essential step for ensuring high-quality assurance standards. This requirement is of primary importance for online vision-based automated pellet stacking system. During nuclear fuel pin manufacturing, the image of a single row (consecutively placed components with no gap) is processed and analyzed to extract meaningful edges. Generally, these edges follow a regular pattern; however, the presence of surface cracks and chips can alter this pattern. In this paper, we formalize the detection of defective parts as a pattern matching problem. Three different patterns are proposed and evaluated for sensitivity, specificity, and accuracy. An experiment performed with the proposed pattern matching techniques show that multi-pattern matching is the most effective method for identifying defective parts.


Multi-pattern matching Aho–Corasick algorithm Finite state automata Fuel pellet Stack Defects 


  1. 1.
    Navarro, G., Raffinot, M.: Flexible pattern matching in strings. Cambridge University Press (2002)Google Scholar
  2. 2.
    Knuth, D.E., Morris Jr, J.H., Pratt, V.R.: Fast pattern matching in strings. SIAM J. Comput. (1997)Google Scholar
  3. 3.
    Boyer, R., Moore, S.: A fast string searching algorithm. Commun. ACM 20(10), 762–772 (1977)CrossRefGoogle Scholar
  4. 4.
    Karp, R., Rabin, M.: Efficient randomized pattern-matching algorithms. IBM J. Res. Dev. 31(2), 249–260 (1987)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Yao, A.C.: The complexity of pattern matching for a random string. SIAM J. Comput. 368–387 (1979)Google Scholar
  6. 6.
    Tran, N-P., Lee, M, Hong, S., Choi J.: High throughput parallel implementation of Aho-Corasick algorithm on a GPU. In: IEEE 27th International Symposium on Parallel and Distributed Processing Workshops and Ph.D. Forum (2013)Google Scholar
  7. 7.
    Kim, Hwee, Han, Yo-Sub: OMPPM: online multiple palindrome pattern matching. Bioinformatics 32(8), 1151–1157 (2016)CrossRefGoogle Scholar
  8. 8.
    Lin, P-C., Li, Z-X., Lin, Y-D.: Profiling and accelerating string matching algorithms in three network content security applications. In: IEEE Communications Surveys 2nd Quarter, vol. 8, no. 2 (2006)Google Scholar
  9. 9.
    Sharma, S., Jaju, A., Singh, N., Pal, P.K., Raju, Y.S., Rama Krishna Murthy, G.V.: Robotic system for stacking PHWR nuclear fuel pellets using machine vision. In: Characterization and Quality Control of Nuclear Fuels (CQCNF-2013) (2013)Google Scholar
  10. 10.
    Singh, N., Jaju, A., Sharma, S., Pal, P.K.: Online vision-based measurement of stacks of nuclear fuel pellets in a tray. In: Advances in Robotics Conference (2015)Google Scholar
  11. 11.
    Bulnes, F.G., Usamentiaga, R., Garcia, D.F., Molleda, J.: Detection of periodical patterns in the defects identified by computer vision systems. In: 11th International Conference on Intelligent Systems Design and Applications (2011)Google Scholar
  12. 12.
    Dori, S., Landau, G.M.: Construction of Aho Corasick automaton in linear time for integer alphabets. Elsevier (2006)Google Scholar
  13. 13.
    Altman, D.G., Bland, J.M.: Diagnostic tests. 1: Sensitivity and specificity. BMJ 308(6943), 1552 (1994)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.Division of Remote Handling and RoboticsBhabha Atomic Research CentreMumbaiIndia

Personalised recommendations