An Improved Mean Shift Algorithm for Vehicle Tracking

Abstract:

Classical mean shift tracking algorithm doesnt show good performance when the tracked objects move fast, change in size or pose. This paper proposes an improved mean shift method used for vehicle tracking. Firstly, a position prediction model based on second order auto-regression process is used to find the initial position of mean shift iteration, reduce times of iteration and enhance the tracking accuracy. Secondly, we employ a position search method based on the weight image to improve the tracking result when the result of basic mean shift tracking is not good. The proposed algorithm is tested in a real traffic video to track a vehicle changing in size and pose with more accurate result than basic mean shift tracking algorithm.

• Periodical: Advanced Materials Research (Volumes 718-720)
• Pages: 2329-2334
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.718-720.2329
• Online since: July 2013
• Authors: Yan Shuang Hao, Yi Xin Yin, Jin Hui Lan, Shu Wei Xiao
• Keywords: Mean-Shift, Position Prediction, Vehicle Tracking, Weight Image

References

[1] ZOU Xiaohong, LI Dongmei, LIU Jichuan, Real-time vehicles tracking based on Kalman filter in an ITS, International Symposium on Photoelectronic Detection and Imaging 2007: Image Processing. 6623 (2008) 1-7.

DOI: 10.1117/12.791270

[2] Jianguang Lou, Tieniu Tan, Weiming Hu, et al, 3-D Model-Based Vehicle Tracking, IEEE Transaction on Image Processing. 14 (2005) 1561-1569.

DOI: 10.1109/tip.2005.854495

[3] M.K. Pitt, N. Shephard, Filtering via simulation: auxiliary particle filters, Journal of the American Statistical Association. 94 (1999) 590–599.

DOI: 10.1080/01621459.1999.10474153

[4] Jacob Scharcanski, Alessandro Bof de Oliveira, et al, A Particle-Filtering Approach for Vehicular Tracking Adaptive to Occlusions, IEEE Transaction on Vehicular Technology. 60 (2011) 381-389.

DOI: 10.1109/tvt.2010.2099676

[5] Wei Leong Khong, Wei Yeang Kow, Farrah Wong, et al, Enhancement of Particle Filter Approach for Vehicle Tracking via Adaptive Resampling Algorithm, IEEE International Conference on Computational Intelligence, Communication Systems and Networks. (2011) 259-263.

DOI: 10.1109/cicsyn.2011.62

[6] K. Fukunaga, L.D. Hostetler, The estimation of the gradient of a density function with applications in pattern recognition, IEEE Transaction on Information Theory. 21 (1975) 32-40.

DOI: 10.1109/tit.1975.1055330

[7] Yizong Cheng, Mean Shift, Mode Seeking, and Clustering, IEEE Transaction on Pattern Analysis and Machine Intelligence. 17 (1995) 790-799.

DOI: 10.1109/34.400568

[8] D. Comaniciu, V. Ramesh, et al, Kernel-Based Object Tracking, IEEE Transaction on Pattern Analysis and Machine Intelligence. 25 (2003) 564-577.

DOI: 10.1109/tpami.2003.1195991

[9] PENG Ning-Song, YANG Jie, LIU Zhi, ZHANG Feng-Chao, Automatic Selection of Kernel-Bandwidth for Mean-Shift Object Tracking, Journal of Software. 16 (2005) 1542-1550.

DOI: 10.1360/jos161542

[10] JIA Jing-ping, CHAI Yan-mei, ZHAO Rong-chun, Robust Tracking of Objects in Image Sequences Using Multiple Degrees of Freedom Mean Shift Algorithm, Journal of Image and Graphics. 11 (2006) 707-713.

DOI: 10.1109/icmlc.2005.1527848

[11] J. Ning, L. Zhang, D. Zhan, C. Wu, Scale and orientation adaptive mean shift tracking, IET Computer Vision. 6 (2012) 52-61.

DOI: 10.1049/iet-cvi.2010.0112

[12] CHANG Fa-liang, LIU Xue, WANG Hua-jie, Target tracking algorithm based on mean shift and Kalman filter, Computer Engineering and Applications. 43 (2007) 50-52.

[13] LI Wen-hui, ZHOU Qiang, WANG Ying, ZHANG De-cai, Adaptive tracking algorithm based on particle filter-mean shift, Journal of Jilin University (Engineering and Technology Edition). 42 (2012) 407-411.