access icon free Robust video tracking algorithm: a multi-feature fusion approach

© The Institution of Engineering and Technology
Received 21/08/2017, Accepted 13/02/2018, Revised 28/12/2017, Published 14/02/2018

This study proposes a novel robust video tracking algorithm consists of target detection, multi-feature fusion, and extended Camshift. Firstly, a novel target detection method that integrates Canny edge operator, three-frame difference, and improved Gaussian mixture model (IGMM)-based background modelling is provided to detect targets. The IGMM-based background modelling divides video frames into meshes to avoid pixel-wise processing. In addition, the output of the target detection is utilised to initialise the IGMM and to accelerate the convergence of iterations. Secondly, low-dimensional regional covariance matrices are introduced to describe video targets by fusing multiple features like pixel location, colour index, rotation and scale invariant features as well as uniform local binary patterns, and directional derivatives. Thirdly, an extended Camshift based on adaptive kernel bandwidth and robust H ∞ state estimation is proposed to predict the states of fast moving targets and to reduce the mean shift iterations. Finally, the effectiveness of the proposed tracking algorithm is demonstrated via experiments.

Inspec keywords: edge detection; image fusion; Gaussian processes; mixture models; covariance matrices; video signal processing; feature extraction

Other keywords: robust H∞ state estimation; extended Camshift; multifeature fusion approach; local binary patterns; low-dimensional regional covariance matrices; target detection method; multiple feature fusion; video targets; robust video tracking algorithm; directional derivatives; target detection; adaptive kernel bandwidth; improved Gaussian mixture model; tracking algorithm; scale invariant features; colour index; pixel location; IGMM-based background modelling; Canny edge operator; video frames; three-frame difference; mean shift iteration reduction

Subjects: Video signal processing; Sensor fusion; Algebra; Image recognition; Other topics in statistics; Other topics in statistics; Computer vision and image processing techniques; Algebra

References

    1. 1)
      • 10. Canny, J.: ‘A computational approach to edge detection’, IEEE Trans. Pattern Anal. Mach. Intell., 1986, 8, (6), pp. 679–698.
    2. 2)
      • 14. Tuzel, O., Porikli, F., Meer, P.: ‘Region covariance: a fast descriptor for detection and classification’. Proc. Int. Conf. Computer Vision, Amsterdam, The Netherlands, 2006, vol. 3952, pp. 589–600.
    3. 3)
      • 22. Yang, H.H.: ‘Research on the detection and tracking of moving target base on Kernel method’, Int. J. Grid Distrib. Comput., 2015, 8, (2), pp. 91–100.
    4. 4)
      • 2. Kim, G.W., Kang, D.S.: ‘Improved Camshift algorithm based on Kalman filter’, Adv. Sci. Technol. Lett., 2015, 98, pp. 135–137.
    5. 5)
      • 24. Dou, J.F., Li, J.X.: ‘Robust visual tracking based on joint multi-feature histogram by integrating particle filter and mean shift’, Optik, 2015, 126, pp. 1449–1456.
    6. 6)
      • 27. Saat, S., Nguang, S.K.: ‘Nonlinear H∞ output feedback control with integrator for polynomial discrete-time systems’, Int. J. Robust Nonlinear Control, 2015, 25, (7), pp. 1051–1065.
    7. 7)
      • 18. Hong, X., Zhao, G., Pietikäinen, M., et al: ‘Combining LBP difference and feature correlation for texture description’, IEEE Trans. Image Process., 2014, 23, (6), pp. 2557–2568.
    8. 8)
      • 5. Lee, L.K., An, S.Y.: ‘Robust visual object tracking with extended Camshift in complex environments’. Proc. Int. Conf. IEEE Industrial Electronics Society, 2011, pp. 4536–4542.
    9. 9)
      • 7. Yan, Z.G., Liang, W.G., Lv, H.D.: ‘A target tracking algorithm based on improved Camshift and UKF’, J. Softw. Eng. Appl., 2014, 7, pp. 1065–1073.
    10. 10)
      • 13. Bhosle, P.I., Bijole, L.: ‘Video object segmentation and tracking using GMM and GMM-RBF method for surveillance system’, Int. J. Recent Innov. Trends Comput. Commun., 2015, 3, (6), pp. 3588–3594.
    11. 11)
      • 26. Simon, D.: ‘Optimal state estimation Kalman, H∞ and nonlinear approaches’ (John Wiley & Sons, Inc., Publication, Australia, 2006).
    12. 12)
      • 3. Hsia, C.H., Liou, Y.J., Chiang, J.S.: ‘Directional prediction Camshift algorithm based on adaptive search pattern for moving object tracking’, J. Real-Time Image Process., 2016, 12, (1), pp. 183–195.
    13. 13)
      • 1. Exner, D., Bruns, E., Kurz, D., et al: ‘Fast and robust Camshift tracking’. Proc. Int. Conf. Computer Vision and Pattern Recognition, San Francisco, CA, 2010, pp. 9–16.
    14. 14)
      • 25. Wang, H., Nguang, S.K.: ‘Multi-target video tracking based on improved data association and mixed Kalman/H∞ filtering’, IEEE Sens. J., 2016, 15, (21), pp. 7693–7704.
    15. 15)
      • 17. Xi, M., Chen, L., Polajnar, D., et al: ‘Local binary pattern network: a deep learning approach for face recognition’. Proc. Int. Conf. Image Processing, Phoenix, USA, 2016, pp. 3224–3228.
    16. 16)
      • 12. Bouwmans, T.: ‘Traditional and recent approaches in background modelling for foreground detection: an overview’, Comput. Sci. Rev., 2014, 11, pp. 31–36.
    17. 17)
      • 11. Varadarajan, S., Miller, P., Zhou, H.Y.: ‘Region-based mixture of Gaussians modelling for foreground detection in dynamic scenes’, Pattern Recognit., 2015, 48, (11), pp. 3488–3503.
    18. 18)
      • 28. Wu, Y., Lim, J., Yang, M.H.: ‘Object tracking benchmark’, IEEE Trans. Pattern Anal. Mach. Intell., 2015, 37, (9), pp. 1834–1848.
    19. 19)
      • 16. Wang, H., Nguang, S.K.: ‘Robust video target tracking based on multi-feature fusion and H∞ filtering’, Int. J. Comput. Commun. Eng., 2016, 5, (22), pp. 79–97.
    20. 20)
      • 21. Ning, J.F., Zhang, L., Zhang, D.: ‘Scale and orientation adaptive mean shift tracking’, IET Comput. Vis., 2012, 6, (1), pp. 52–61.
    21. 21)
      • 19. Pietikäinen, M., Zhao, G.: ‘Two decades of local binary patterns: a survey’, in ‘Advances in independent component analysis and learning machines’, (Academic Press, London, 2015), pp. 175–210.
    22. 22)
      • 15. Förstner, W., Moonen, B.: ‘A metric for covariance matrices’. Proc. Int. Conf. Challenge of the 3rd Millennium, 2003, pp. 299–309.
    23. 23)
      • 8. Zhang, T., Xu, C.S., Yang, M.H.: ‘Multi-task correlation particle filter for robust object tracking’. Proc. Int. Conf. on Computer Vision and Pattern Recognition, 2017, pp. 4335–4343.
    24. 24)
      • 4. Wang, Q.X., Li, X.: ‘A New improved Camshift algorithm and its experimental analysis’. Int. Conf. on Electronic Science and Automation Control, 2015, pp. 275–280.
    25. 25)
      • 23. Anand, S., Mittal, S., Tuzel, O.: ‘Semi-supervised Kernel mean shift clustering’, IEEE Trans. Pattern Anal. Mach. Intell., 2014, 36, (6), pp. 1201–1215.
    26. 26)
      • 29. Andrews, S.: ‘BGSLibrary: an openCV C + + background subtraction library’. Proc. Int. IX Workshop de Visão Computacional, Rio de Janeiro, Brazil, 2013.
    27. 27)
      • 20. Zhao, G., Ahonen, T., Matas, J., et al: ‘Rotation-invariant image and video description with local binary pattern features’, IEEE Trans. Image Process., 2011, 21, (4), pp. 1465–1477.
    28. 28)
      • 6. Zhang, Y.Y., Zhao, X.M., Li, F.j.: ‘Robust object tracking based on simplified codebook masked Camshift algorithm’. Mathematical Problems in Engineering, 2015, pp. 1–12.
    29. 29)
      • 9. Zhang, T., Liu, S., Xu, C., et al: ‘Mining semantic context information for intelligent video surveillance of traffic scenes’, IEEE Trans. Ind. Inf., 2013, 9, (1), pp. 149–160.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cvi.2017.0404
Loading

Related content

content/journals/10.1049/iet-cvi.2017.0404
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading