Video Segmentation Framework Based on Multi-kernel Representations and Feature Relevance Analysis for Object Classification

  • S. Molina-Giraldo
  • J. Carvajal-González
  • A. M. Álvarez-Meza
  • G. Castellanos-Domínguez
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 318)


A video segmentation framework to automatically detect moving objects in a scene using static cameras is proposed. Using Multiple Kernel Representations, we aim to enhance the data separability into the scene by incorporating multiple information sources into the process, and employing a relevance analysis each source is automatically weighted. A tuned Kmeans technique is employed to group pixels as static or moving objects. Moreover, the proposed methodology is tested for the classification of people and abandoned objects. Attained results over real-world datasets, show how our approach is stable using the same parameters for all experiments.


Background subtraction Multiple kernel learning Relevance analysis Data separability 



This research was carried out under grants provided by a M.Sc. and a Ph.D. scholarship provided by Universidad Nacional de Colombia, and the project 15,795, funded by Universidad Nacional de Colombia.


  1. 1.
    Chen, T.W., Hsu, S.C., Chien, S.Y.: Robust video object segmentation based on k-means background clustering and watershed in ill-conditioned surveillance systems. In: IEEE International Conference on Multimedia and Expo, pp. 787–790 (2007)Google Scholar
  2. 2.
    González, J.C., Álvarez-Meza, A., Castellanos-Domínguez, G.: Feature selection by relevance analysis for abandoned object classification. In: CIARP, pp. 837–844 (2012)Google Scholar
  3. 3.
    Klare, B., Sarkar, S.: Background subtraction in varying illuminations using an ensemble based on an enlarged feature set. In: Computer Vision and Pattern Recognition Workshops, IEEE Computer Society Conference on CVPR Workshops, pp. 66–73 (2009)Google Scholar
  4. 4.
    Elgammal, A., Duraiswami, R., Harwood, D., Davis, L.: Background and foreground modeling using nonparametric kernel density estimation for visual surveillance. Proc. IEEE 90, 1151–1163 (2002)CrossRefGoogle Scholar
  5. 5.
    Maddalena, L., Petrosino, A.: A self-organizing approach to background subtraction for visual surveillance applications. IEEE Trans. Image Process. 17, 1168–1177 (2008)CrossRefMathSciNetGoogle Scholar
  6. 6.
    Gutchess, D., Trajkovics, M., Cohen-Solal, E., Lyons, D., Jain, A.: A background model initialization algorithm for video surveillance. In: Computer Vision, 2001. ICCV 2001. Proceedings. Eighth IEEE International Conference on IEEE, vol. 1, pp. 733–740 (2001)Google Scholar
  7. 7.
    Rakotomamonjy, A., Bach, F.R., Canu, S., Grandvalet, Y.: SimpleMKL. J. Mach. Learn. Res. 9, 2491–2521 (2008)MathSciNetMATHGoogle Scholar
  8. 8.
    Gonen, M., Alpaydin, E.: Localized multiple kernel regression. In: Proceedings of the 20th International Conference on Pattern Recognition (ICPR) (2010)Google Scholar
  9. 9.
    Daza-Santacoloma, G., Arias-Londoo, J.D., Godino-Llorente, J.I., Senz-Lechn, N., Osma-Ruz, V., Castellanos-Domnguez, G.: Dynamic feature extraction: an application to voice pathology detection. Intell. Autom. Soft Comput. 15(4), 665–680 (2009)Google Scholar
  10. 10.
    Cuesta-Frau, D., Pérez-Cortés, J., Andreu-Garcia, G.: Clustering of electrocardiograph signals in computer-aided holter analysis. Comput. Methods Programs Biomed. 72, 179–196 (2003)CrossRefGoogle Scholar
  11. 11.
    Raty, T.: Survey on contemporary remote surveillance systems for public safety. IEEE Trans. Man, Cybern., Part C: Appl. Rev. 40, 493–515 (2010)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • S. Molina-Giraldo
    • 1
  • J. Carvajal-González
    • 1
  • A. M. Álvarez-Meza
    • 1
  • G. Castellanos-Domínguez
    • 1
  1. 1.Signal Processing and Recognition GroupUniversidad Nacional de ColombiaManizalesColombia

Personalised recommendations