Abstract
Mean shift algorithm has attracted much attention in computer vision and has recently shown promising performance in the challenging problem of visual tracking, but it is difficult to deal with occlusion. In this paper, a timely occlusion object detection based on mean shift is proposed. By analyzing occlusion process, it is evident to find that occluded size is increasing and occlusion patch lies to edge of objects at the beginning. So object model is divided into several parts. In order to reduce computation, only edge patches is considered. If Bhattacharyya coefficient of one patch decreases greatly and other patches change faintly, it means that object is occluded in this area. In this method, object model is divided into two parts, one is occlusion part, the other is no occlusion part, the no occlusion part of the object can be obtained to track object continuatively until object is occluded totally. Experiments show that compared with whole object model judges, it is timely to detect occlusion and deals with occlusion successfully.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ying, M., Jingjue, J.: Detection and tracking of moving objects under occlusion and shadow. In: Proc. of SPIE, vol. 6044 (2005)
Comaniciu, D., Ramesh, V., Meer, P.: Kernel-Based Object Tracking. IEEE Trans. on Pattern Analysis and Machine Intelligence 25(5), 564–577 (2003)
Peng, N., Yang, J., Liu, E.: Model update mechanism for mean-shift tracking. Journal of Systems Engineering and Electronic 16(1), 52–57 (2005)
Fan, Z., Wu, Y., Yang, M.: Multiple collaborative kernel tracking. In: Proc. of CVPR, pp. 502–509 (2005)
Collins, R.T.: Mean-Shift Blob Tracking through Scale Space. In: 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2003, vol. 2, pp. 234–240. IEEE, Vancouver (2003)
Li, Z., Tang, Q.L., Sang, N.: Improved mean shift algorithm for occlusion pedestrian tracking. Electronics Letters 44(10) (2008)
Li, Z.: Improved mean shift algorithm for multiple occlusion target tracking. Optical Engineering 47(8) (2008)
Senior, A., Hampapur, A., Tian, Y.L., Brown, L., Pankanti, S., Bolle, R.: Appearance models for occlusion handling. In: 2nd Intl. Workshop Performance Eval. Track. Surveill. Syst. (2001)
Kaucic, R., Perera, A.G., Brooksby, G., Kaufhold, J., Hoogs, A.: A unified framework for tracking through occlusions and across sensor gaps. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 990–997. IEEE (2005)
Amitha Perera, A.G., Srinicas, C., Hoogs, A., Brooksby, G., Hu, W.: Multiobject tracking through simultaneous long occlusions and split-merge. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 666–673. IEEE (2006)
Han, M., Xu, W., Tao, H., Gong, Y.: An algorithm for multiple object trajectory tracking. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 1864–1871. IEEE (2004)
Loutas, E., Nikou, C., Pitas, I.: Information theory-based analysis of partial and total occlusion in object tracking. In: Proceedings 2002 International Conference on, vol. 2 (2002)
Zou, X., Li, D., Liu, J.: Real-time vehicles tracking based on Kalman filter in an ITS. In: Proc. of SPIE, vol. 6623 (2008)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chen, Ah., Yang, Bq., Chen, Zg. (2012). A Timely Occlusion Detection Based on Mean Shift Algorithm. In: Deng, W. (eds) Future Control and Automation. Lecture Notes in Electrical Engineering, vol 173. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-31003-4_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-31003-4_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-31002-7
Online ISBN: 978-3-642-31003-4
eBook Packages: EngineeringEngineering (R0)