Sliding window kernel ridge regression trajectory predicting algorithm

Zhao Fei; Lu Huanzhang; Zhang Zhiyong

Due to the requirement for prediction of nonlinear target trajectory in image sequences, a sliding window kernel ridge regression(KRR) target trajectory predicting algorithm was proposed. The full KRR which posses the bias item was deduced first, and then the iterative form of sliding window KRR algorithm was also derived in this paper. The algorithm was carried out in a sliding way, and the trajectory information in latest frames was used to predict the position in the next frame, which was achieved by using the KRR. The experimental results demonstrate that the proposed algorithm can predict the nonlinear trajectories accurately, and the prediction error is small. The structure of the proposed algorithm is simple and practicable in engineering application.

1. National Key Laboratory of Automatic Target Recognition ATR,National University of Defense Technology,Changsha 410073,China

Received Date: 2012-07-23

Rev Recd Date: 2012-08-29

Publish Date: 2013-03-25

Key words:

Abstract: Due to the requirement for prediction of nonlinear target trajectory in image sequences, a sliding window kernel ridge regression(KRR) target trajectory predicting algorithm was proposed. The full KRR which posses the bias item was deduced first, and then the iterative form of sliding window KRR algorithm was also derived in this paper. The algorithm was carried out in a sliding way, and the trajectory information in latest frames was used to predict the position in the next frame, which was achieved by using the KRR. The experimental results demonstrate that the proposed algorithm can predict the nonlinear trajectories accurately, and the prediction error is small. The structure of the proposed algorithm is simple and practicable in engineering application.

HTML

Reference (23)

[1]	Liu Fusheng, Luo Pengfei. Statistical Signal Processing [M]. Changsha: National University of Defense Technology Press, 1999: 196-208. (in Chinese) 刘福声, 罗鹏飞. 统计信号处理[M]. 长沙: 国防科技大学出版社, 1999: 196-208.
[2]
[3]	Gong Yaxin, Yang Hongwen, Hu Weidong, et al. Multiple model particle filter based track -before -detect for maneuvering weak target [J]. Journal of Electronics Information Technology, 2008, 30 (4): 941 -944. (in Chinese) 龚亚信, 杨宏文, 胡卫东, 等. 基于多模粒子滤波的机动弱目标检测前跟踪[J]. 电子与信息学报, 2008, 30(4): 941-944.
[4]
[5]
[6]	Li Shaojun, Zhu Zhenfu. Track before detect algorithm via particle filtering [J]. Infrared and Laser Engineering, 2009, 38(2): 352-357. (in Chinese) 李少军, 朱振福. 采用粒子滤波的先跟踪后检测算法[J]. 红外与激光工程, 2009, 38(2): 352-357.
[7]
[8]	Wang Wanping, Ren Qifeng, Liao Sheng, et al. Application of particle filtering in single moving platform passive ranging[J]. Infrared and Laser Engineering, 2009, 38(4): 716-720. (in Chinese) 王万平, 任栖锋, 廖胜, 等. 粒子滤波器在运动单站被动测距中的应用[J]. 红外与激光工程, 2009, 38(4): 716-720.
[9]
[10]	Zhang Bing, Lu Huanzhang. The detection algorithm for moving point target rrajectory in image sequences [J]. Aata Electronica Sinca, 2004, 9(9): 1524-1526. (in Chinese) 张兵, 卢焕章. 序列图像中运动点目标轨迹检测算法研究[J]. 电子学报, 2004, 9(9): 1524-1526.
[11]
[12]	Zhang Lu. Research on some key techniques for image homing guidance based on strap -down platform [D]. Changsha: National University of Defense Technology, 2010. (in Chinese) 张路. 捷联平台成像末制导关键技术研究[D]. 长沙: 国防科学技术大学, 2010.
[13]
[14]	Xiaoying Deng, Yong Luo, Tao Liu, et al. Least squares support vector regression filter [C]//2010 3rd International Congress on Image and Signal Processing, 2010: 730-733.
[15]	Fernando PerezCruz, Sanjeev R Kulkarni. Robust and low complexity distributed kernel least squares learning in sensor networks [J]. IEEE Singal Processing Letters, 2010, 17(4): 355-358.
[16]
[17]	Weifeng Liu, Pokharel P P, Principe J C. The kernel least -mean -square algorithm [J]. IEEE Transactions on Signal Processing, 2008, 56(2): 543-554.
[18]
[19]	Cedric Richard, Jose Carlos M Bermudez, Paul Honeine. Online prediction of time series data with kernels [J]. IEEE Transactions on Signal Processing, 2009, 57(3): 1058-1067.
[20]
[21]	Fu Huixuan, Sun Feng, Liu Sheng. Anti-occlusion tracking algorithm based on LSSVM prediction and Kalman -mean shift [C]//Proceedings of the 8th World Congress on Intelligent Control and Automation, 2010: 6031-6036.
[22]
[23]	Saunders C, Gammerman A, Vovk A. Ridge regression learning algorithm in dual variables [C]//Proceedings of the 15th International Conference on Machine Learning, 1998: 515-521.

Sliding window kernel ridge regression trajectory predicting algorithm

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Related

Proportional views