Advertisement

Image Classification Using Legendre–Fourier Moments and Artificial Neural Network

  • Abderrahmane MachhourEmail author
  • Mostafa El Mallahi
  • Zakia Lakhliai
  • Ahmed Tahiri
  • Driss Chenouni
Conference paper
  • 118 Downloads
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1076)

Abstract

The nonlinear structure of the artificial neural network is efficient for the classification; however, the choice of features is a fundamental problem due to their direct impact on the network convergence and performance. In this paper, we present a new method of image classification method based on Legendre–Fourier moments using an artificial neural network. We used LFMs to extract features from images. In result, every image is represented by a descriptor vector; these vectors are inputs of our neural network. We tested this model on Fashion-MNIST database and we got important results; the model’s accuracy exceeds 97%. The validity of this proposed method has provided under different transformations.

Keywords

Image classification Legendre–Fourier moments Artificial neural network Features Descriptor vector Moment invariants 

References

  1. 1.
    Amakdouf, H., El Mallahi, M., Zouhri, A., Qjidaa, H.: Classification and recognition of 3D image of Charlier moments using multilayer perceptron architecture. Procedia Comput. Sci. 127(2018), 226–235 (2018)CrossRefGoogle Scholar
  2. 2.
    Hu, M.K.: Visual pattern recognition by moment invariants. IRE Trans. Inf. Theory 8(2), 179–187 (1962)CrossRefGoogle Scholar
  3. 3.
    Teague, M.R.: Image analysis via the general theory of moments. J. Opt. Soc. Am. 70(8), 920–930 (1980)MathSciNetCrossRefGoogle Scholar
  4. 4.
    El Mallahi, M., Zouhri, A., Amakdouf, H., Qjidaa, H.: Rotation scaling and translation invariants of 3D radial shifted Legendre moments. Int. J. Autom Comput. 15(2), 169–180 (2018)Google Scholar
  5. 5.
    Sheng, Y., Shen, L.: Orthogonal Fourier-Mellin moments for invariant pattern recognition. JOSA A 11(6), 1748–1757 (1994)CrossRefGoogle Scholar
  6. 6.
    Ping, Z., Wu, R., Sheng, Y.: Image description with Chebyshev-Fourier moments. J. Opt. Soc. Am., A Opt. Image Sci. Vis. 19(9), 1748–54 (2002)Google Scholar
  7. 7.
    Yang, B., Dai, M.: Image analysis by Gaussian-Hermite moments. Signal Process 91(10), 2290–2303 (2011)CrossRefGoogle Scholar
  8. 8.
    Hosny, K.M.: Image representation using accurate orthogonal Gegenbauer moments. Pattern Recognit. Lett. 32(6), 795–804 (2011)CrossRefGoogle Scholar
  9. 9.
    El Mallahi, M., Mesbah, A., El Fadili, H., Zenkouar, K., Qjidaa, H.: Compact computation of Krawtchouk moments for 3D object representation. WSEAS Trans. Circ. Syst. 13. E-ISSN: 2224-266X (2014)Google Scholar
  10. 10.
    El Mallahi, M., Zouhri, A., El-Mekkaoui, J., Qjidaa, H.: Three dimensional radial Krawtchouk moment invariants for volumetric image recognition. Pattern Recogn. Image Anal. 27(4), 810–824 (2017)Google Scholar
  11. 11.
    El Mallahi, M., Mesbah, A., Qjidaa, H.: Fast algorithm for 3D local feature extraction using Hahn and Charlier moments. Advances in Ubiquitous Networking 2. Lecture Notes in Electrical Engineering, vol. 397. Springer, Singapore (2016)Google Scholar
  12. 12.
    Mesbah, A., El Mallahi, M., Qjidaa, H.: Fast and efficient computation of three-dimensional Hahn moments. SPIE, J. Electron. Imaging. 25(6), 061621 (2016)Google Scholar
  13. 13.
    El Mallahi, M., Mesbah, A., Qjidaa, H.: 3D radial invariant of dual Hahn moments. Neural Comput. Appl. 30(7), 2283–2294 (2018)Google Scholar
  14. 14.
    Flusser, J., Suk, T., Zitová, B.: Moments and Moment Invariants in Pattern Recognition. Wiley, Chichester (2009)CrossRefGoogle Scholar
  15. 15.
    El Mallahi, M., Zouhri, A., Mesbah, A., El Affar, I., Qjidaa, H.: Radial invariant of 2D and 3D Racah moments. Multimedia Tools Appl. 77(6), pp. 6583–6604 (2018)Google Scholar
  16. 16.
    El Mallahi, M., Zouhri, A., Mekkaoui, J., Qjidaa, H.: Radial Meixner moments for rotational invariant pattern recognition. IEEE 2017 Intelligent Systems and Computer Vision, ISCV (2017)Google Scholar
  17. 17.
    El Mallahi, M., Zouhri, A., Qjidaa, H.: Radial Meixner moment invariants for 2D and 3D image recognition. Pattern Recogn. Image Anal. 28(2), 207–216 (2018)CrossRefGoogle Scholar
  18. 18.
    El Mallahi, M., Mesbah, A., Qjidaa, H.: Radial Charlier moment invariants for 2D object/image recognition. IEEE, International Conference on Multimedia Computing and Systems -Proceedings, Procedia Computer Science, vol. 127, pp. 226–235 (2018)Google Scholar
  19. 19.
    Xin, Y., Pawlak, M., Liao, S.: Accurate computation of Zernike moments in polar coordinates. IEEE Trans. Image Process. 16(2), 581–587 (2007)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  • Abderrahmane Machhour
    • 1
    Email author
  • Mostafa El Mallahi
    • 1
  • Zakia Lakhliai
    • 1
  • Ahmed Tahiri
    • 1
  • Driss Chenouni
    • 1
  1. 1.Sidi Mohamed Ben Abdellah University, Laboratory of Computer Science and Interdisciplinary Physics LIPI, ENSFezMorocco

Personalised recommendations