Research on Kalman Particle Filter-Based Tracking Algorithm

Abstract:

In the application of computer vision technique, target tracking in image sequences was an important research subject. This paper describes the particle filter and introduces a tracking algorithm based on Kalman particle filter. The algorithm improves the traditional particle filter, whose non-linear and non-Gaussian may result in non-robustness of tracking process. Kalman particle filter use kalman filter to predict the particle’s state and generate the proposal distribution, the state of each particle evolved by the Kalman prediction equations and update equations, increasing the robustness of tracking. Experimental results show that the proposed method in comparison with the traditional particle filtering can be more accurate on tracking and ensure the robustness of performance in a complex environment.

• Periodical: Advanced Materials Research (Volume 461)
• Pages: 571-574
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.461.571
• Online since: February 2012
• Authors: Yi Min Hou, Yong Liang Zhao, Ting Ting Sun, Jian Ming Di
• Keywords: Kalman Filter (KF), Nonlinear, Object Tracking, Particle Filter (PF)

References

[1] M.S. Grewal, A.P. Andrews, Kalman Filtering: Theory and Practice Using MATLAB, John Wiley & Sons (2001).

[2] B.D. O Anderson, J.B. Moore, Optimal filtering, Englewood Cliffs (1979).

[3] M. Isard, A. Blake, Contour tracking by stochastic propagation of conditional density, Proceedings of the Fourth European Conference on Computer Vision, 334-356 (1996).

DOI: 10.1007/bfb0015549

[4] M. Isard, A. Blake Condensation-conditional density propagation for visual tracking. International Journal of Computer Vision, 29(1), 5-28 (1996).

[5] M.S. Arulampalam, S. Maskell, N. Gordon, A tutorial on particle filters for on line nonlinear or non-Gaussian Bayesian tracking, IEEE Trans. on Signal Processing, 50(2), 174-188 (2002).

DOI: 10.1109/78.978374

[6] P. Brasnett, L. Mihaylova, D. Bull. Sequential Monte Carlo tracking by fusing multiple cues in video sequences, Image and Vision Computing, 25(8): 1217-1227 (2007).

DOI: 10.1016/j.imavis.2006.07.017

[7] S.K. Zhou, R. Chellappa, B. Moghaddam, Visual tracking and recognition using appearance-adaptive models in particle filters, IEEE Trans. on Image Process, 13(11), 1491-1506 (2004).

DOI: 10.1109/tip.2004.836152

[8] C. Li, C. Xu, C. Gui, M.D. Fox, Level set evolution without re-initialization, a new variation formulation. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 430-436 (2005).

DOI: 10.1109/cvpr.2005.213

[9] M. Isard, A. Blake, Condensation conditional density propagation for visual tracking, International Journal of Computer Vision, 29 (1), 28-52 (1998).

[10] J.F.G. Freitas, M. Niranjag, A.H. Gee, Sequential Monte Carlo methods to train neural networks models, Neural Computer, 12(4), 955-993 (2000).

DOI: 10.1162/089976600300015664

[11] H.W. Shi, Human motion tracking based on statistical model and genetic particle filter. Application Research of Computers, 25 (4), 109-210 (2001).

[12] T. X Yuan, Principles of best estimate. National Defence Industry Press (1980).

[13] X.R. Li, V.P. Jilkov, A Survey of maneuvering target tracking-part I: dynamic models, IEEE Transactions on Aerospace and Electronic Systems, 39(4), 1333-1364 (2003).

DOI: 10.1109/taes.2003.1261132

[14] X.R. Li, V.P. Jilkov, A survey of maneuvering target tracking-Part III: Measurement Models, Proceedings of SPIE Conference on Signal and Data Processing of Small Targets, 423-446 (2001).

DOI: 10.1117/12.492752

[15] M.K. Pitt, N. Shephard: Filtering via simulation, Auxiliary particle filters, 94(466), 590-599 (1999).

DOI: 10.1080/01621459.1999.10474153