Occluded Object Tracking Based on Mean Shift and Accumulation Error Suppression

Abstract:

To handle occlusion and accumulation error in tracking procedure, a novel method is proposed. Firstly, the object feature is modeled with kernel based color histogram. Then, mean shift is used to localizing the object with Kalman filter providing initial iteration location and scale. Object observation value is acquired by weighting the similarities of hue and saturation in x and y-directions. Finally, occlusion and scene disturbance are judged by maximal similarity and the matching deviation, so as to selectively update the object model. To suppress the accumulation error, the noise covariance is updated according to the iteration error in the latest N frames. Experimental results show that the proposed method is robust in tracking the occluded objects under complex scene.

• Periodical: Applied Mechanics and Materials (Volumes 117-119)
• Pages: 20-24
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.117-119.20
• Online since: October 2011
• Authors: Hong Lu, Hong Sheng Li, Shu Min Fei, Wei Feng Cao
• Keywords: Accumulation Error, Mean-Shift, Model Updating, Occluded Object, Tracking

References

[1] G. R. Legters, T. Y. Yong: IEEE Trans on Pattern Analysis and Machine Intelligence, Vol.4 (1982), pp.583-594.

[2] N. P. Papanikolopoulos, P. K. Khosla, T. Kanade: IEEE Trans on Robotics and Automation, Vol.9(1993), pp.14-35.

[3] N. S. Peng, J.Yang, Z. Liu: Pattern Recognition Letters, Vol.26 (2005) , p.605–614.

[4] Z. S. SUN, J. X. SUN, J. Z. SONG, S. QIAO: Optics and Precision Engineering, Vol.15 (2007), pp.267-271. In Chinese.

[5] Y. Chen, S. S. Yu, W. P. Sun, X. P. Chen: Object Tracking Using an Improved Kernel Method. Proc. of Int. Conf. on Embedded Software and Systems, (2008), pp.511-515.

DOI: 10.1109/icess.2008.64

[6] H. J. LI, S. X. LIN, Y. D. ZHANG: J. Computer-aided Design & Computer Graphics, Vol.18 (2006), pp.1645-1651. In Chinese.

[7] B. Wu, R. Nevatia.Tracking of Multiple, Partially Occluded Humans Based on Static Body Part Detection: Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, (2006), pp.951-958.

DOI: 10.1109/cvpr.2006.312

[8] J. L. Tu, H. Tao, T. Huang: Machine Vision and Applications, Vol.20 (2009), p.163–173.

[9] D. Comaniciu, V. Ramesh, P. Meer: IEEE Trans on Pattern Analysis and Machine Intelligence, Vol.25 (2003), pp.564-575.

DOI: 10.1109/tpami.2003.1195991

[10] H. Tao, H. S. Sawhney, R. Kumar, S.Corporation:Dynamic Layer Representation with Applications to Tracking. Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, (2000), pp.134-141.

DOI: 10.1109/cvpr.2000.854760

[11] H. T. Nguyen, M. Worring, R. Boomgaard: Occlusion Robust Adaptive Template Tracking. Proc. of IEEE Int. Conf. on Computer Vision, (2001), pp.678-683.

DOI: 10.1109/iccv.2001.937587