Pedestrian Tracking Based on Improved Particle Filter under Complex Background

Abstract:

When tracking a moving human target,the traditional particle filter algorithm based on color characteristic can't get accurate results in situations like complicated background or frequent brightness change.Due to the problem, this paper put forward a particle filter algorithm based on combination of color characteristics and shape features of the target.Firstly, fusing the above-mentioned two features into the particle filter frame to calculate the particle weights and achieve the human tracking goal through image sequences . The experimental results show that the algorithm can improve the traditional tracking algorithms based on single color feature limitations.And greatly improves the accuracy and effectiveness of the human tracking under complex background

• Periodical: Advanced Materials Research (Volumes 756-759)
• Pages: 4103-4109
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.756-759.4103
• Online since: September 2013
• Authors: Yu Fu, Xiang Long
• Keywords: Color Feature, Component, Joint Similarity, Particle Filter (PF), Shape Feature

References

[1] Yasuno M, Ryousuke S, Yasuda N, et al. Pedestran detection and tracking in far infrared images [C]/ Proc. of IEEE Intelligent Transportation Systems, 2005. USA: IEEE, 2005: 182-187.

DOI: 10.1109/itsc.2005.1520035

[2] Binelli E, Broggi A, Fascioli A, et al. A modular tracking system for far infrared pedestrian recognition [C]/ Proc. of IEEE Intelligent Vehicles Symposium, 2005. USA: IEEE, 2005: 759-764.

DOI: 10.1109/ivs.2005.1505196

[3] Fengliang Xu, Xia Liu, Fujimura K. Pedestrian detection and tracking with night vision [J]. IEEE Trans. on Intelligent Transportation Systems (S1524-9050), 2005, 6(1): 63-71. Systems (S1524-9050), 2005, 6(1): 63-71.

DOI: 10.1109/tits.2004.838222

[4] LI Jin, YU Hong, ZHOU Lu-lu, et al. Human Motion Tracking Based on Quantum Genetic and Unscented Particle Filter [J]. Journal of System Simulation (S1004-731X), 2008, 20(18): 4867-4871.

[5] S J McKenna, H Nait-Charif. Tracking human motion using auxiliary particle filters and iterated likelihood weighting [J]. Image and Vision Computing (S0262-8856), 2007, 25(6): 852-862.

DOI: 10.1016/j.imavis.2006.06.003

[6] Zhu W, Levinson S E Edge orientation-based multiview object recognition [C]/ICPR00 Barcclona: IEEE CS Press. 2000: 1936- (1939).

[7] McInerney T . Terzopou Ios D. T-snakes: Topology adaptive snakes[J]. M ed ical Image A nalysis, 2000, 2: 73- 9L.

[8] NUMMIARO K, KOLLER－MEIER E, GOOL L V．An adaptive color－based particle filter[J]．Image and Vision Computing, 20－03, 21（1）：99－110.

DOI: 10.1016/s0262-8856(02)00129-4

[9] Perez P, Hue C, Vermaak J, et al. Color-based probabilistic tracking [C]/ Proc. of 7th European Conference on Computer Vision, Copenhagen, Denmark. USA: IEEE, 2002, 1: 661-675.

DOI: 10.1007/3-540-47969-4_44

[10] BRADSKI G, KAEBLER A．Learning OpenCV[M]．Beijing: Tsinghua University Press, (2009).