Shape-Based Image Retrieval Using k-Means Clustering and Neural Networks

  • Xiaoliu Chen
  • Imran Shafiq Ahmad
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4872)


Shape is a fundamental image feature and belongs to one of the most important image features used in Content-Based Image Retrieval. This feature alone provides capability to recognize objects and retrieve similar images on the basis of their contents. In this paper, we propose a neural network-based shape retrieval system in which moment invariants and Zernike moments are used to form a feature vector. k-means clustering is used to group correlated and similar images in an image collection into k disjoint clusters whereas neural network is used as a retrieval engine to measure the overall similarity between the query and the candidate images. The neural network in our scheme serves as a classifier such that the moments are input to it and its output is one of the k clusters that has the largest similarity to the query image.


image retrieval shape-based image retrieval k-means clustering moment-invariants Zernike moments 


  1. 1.
    Jain, R.: Sf workshop on visual information management system: Workshop report. In: Storage and Retrieval for Image and Video Databases SPIE , vol. 1908, pp. 198–218 (1993)Google Scholar
  2. 2.
    Loncaric, S.: A survey of shape analysis techniques. Pattern Recognition 31(8), 983–1001 (1998)CrossRefGoogle Scholar
  3. 3.
    Hu, M.K.: Visual pattern recognition by moment invariants. IEEE Transaction on Information Theory 8(2), 179–187 (1962)CrossRefGoogle Scholar
  4. 4.
    Teh, C.-H., Chin, R.: Image analysis by the methods of moments. IEEE Transacfion on Pattern Analysis and Machine Intelligence 10(4), 496–513 (1988)zbMATHCrossRefGoogle Scholar
  5. 5.
    Maitra, S.: Moment invariants. Proceedings of the IEEE 67, 697–699 (1979)CrossRefGoogle Scholar
  6. 6.
    Pavlidis, T.: Survey: A review of algorithms for shape analysis. Computer Graphics Image Processing 7(2), 243–258 (1978)CrossRefMathSciNetGoogle Scholar
  7. 7.
    Dai, M., Baylou, P., Najim, M.: An efficient algorithm for computation of shape moments from run-length codes or chain codes. Pattern Recognition 25(10), 1112–1128 (1992)CrossRefGoogle Scholar
  8. 8.
    Chen, X., Ahmad, I.: Neural network-based shape retrieval using moment invariants and zernike moments. Technical report 06-002, School of Computer Science, University of Windsor (January 2006)Google Scholar
  9. 9.
    Zhang, D.S., Lu, G.: A comparative study of three region shape descriptors. In: Proceedings of the Sixth Digital Image Computing and Applications (DICTA 2002), pp. 86–91 (January 2002)Google Scholar
  10. 10.
    Teague, M.: Image analysis via the general theory of moments. Journal of Optical Society of America 70(8), 920–930 (1980)MathSciNetGoogle Scholar
  11. 11.
    Pitas, I.: Digital Image Processing Algorithms. Prentice-Hall, Englewood Cliffs (1993)Google Scholar
  12. 12.
    Zakaria, M., Vroomen, L., Zsombor-Murray, J., van Kessel, H.: Fast algorithm for the computation of moment invariants. Pattern Recognition 20(6), 639–643 (1987)CrossRefGoogle Scholar
  13. 13.
    Lippmann, R.: An introduction to computing with neural nets. IEEE Acoustics, Speech and Signal Processing Magazine 4(2), 4–22 (1987)Google Scholar
  14. 14.
    URL: Laboratory for engineering man/machine systems LEMS - a large binary image database (2006),
  15. 15.
    Guesebroek, J.M., Burghouts, G., Smeulders, A.: The amsterdam library of object images. International Journal of Computer Vision 61(1), 103–112 (2005)CrossRefGoogle Scholar
  16. 16.
    Ahmad, I.: Image indexing and retrieval using moment invariants. In: Proceedings of the 4th iiWAS), Indonesia, pp. 93–104 (September 2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Xiaoliu Chen
    • 1
  • Imran Shafiq Ahmad
    • 1
  1. 1.School of Computer Science, University of Windsor, Windsor, ON N9B 3P4Canada

Personalised recommendations