Multi-target detection and tracking of video sequence based on Kalman_BP neural network

Qu Shiru; Yang Honghong

To improve the recognition rate and speed of the multi-target detection and tracking in the complex background, a tracking method based on neural network Kalman filter with correction mean square error estimation was proposed. Multi-target detection and tracking of the video sequence were achieved. In this method, first of all, the background was extracted accurately through the inter-frame difference method and multi-target detection was achieved combined with background subtraction method,the detection results were optimized utilizing morphological filtering. Then, Kalman_BP neural network filter was used to predict the position of the moving target. The estimation error of the Kalman filter caused by model changing and noise was mainly reduced with BP neural network, which made the predictive results more accurate. Finally, the fast matching of target was achievid via labeling different targets. Target chain was established by using the characteristics that little change of same goal centroid position and the boundary rectangle between the adjacent frames, which brought about the multi-target tracking. Simulation results show that the algorithm can not only track different scenarios targets, but also count the number of targets and display target trajectory rapidly and stably. Compared with the particle filter and other metheds, tracking is more smooth, thus the reliability of the tracking is improved.

1. School of Automation,Northwestern Polytechnical University,Xi'an 710129,China

Received Date: 2013-01-11

Rev Recd Date: 2013-02-25

Publish Date: 2013-09-25

Key words:

Abstract: To improve the recognition rate and speed of the multi-target detection and tracking in the complex background, a tracking method based on neural network Kalman filter with correction mean square error estimation was proposed. Multi-target detection and tracking of the video sequence were achieved. In this method, first of all, the background was extracted accurately through the inter-frame difference method and multi-target detection was achieved combined with background subtraction method,the detection results were optimized utilizing morphological filtering. Then, Kalman_BP neural network filter was used to predict the position of the moving target. The estimation error of the Kalman filter caused by model changing and noise was mainly reduced with BP neural network, which made the predictive results more accurate. Finally, the fast matching of target was achievid via labeling different targets. Target chain was established by using the characteristics that little change of same goal centroid position and the boundary rectangle between the adjacent frames, which brought about the multi-target tracking. Simulation results show that the algorithm can not only track different scenarios targets, but also count the number of targets and display target trajectory rapidly and stably. Compared with the particle filter and other metheds, tracking is more smooth, thus the reliability of the tracking is improved.

HTML

Reference (19)

[1]
[2]	Li Gang, Zeng Ruili, Lin Ling. Moving target detection in video monitoring system[C]//Proceedings of the 6th World Congress on Intelligent Control and Automation, 2006: 9778-9781.
[3]	Ohta N, Kanatani K. Moving object detection from optical flow with out empirical thresholds[J]. IEICE Transactions on Information and Systems, 1998, E81-D(2): 243-245.
[4]
[5]	Zhang Rong, Cheng Xingwu, Zhuo Hongyan, et al. Detection algorithm for moving infrared point target in the sky[J]. Infrared and Laser Engineering, 2003, 32(5): 472-475. 张蓉, 陈兴无, 卓红艳, 等. 空中红外运动点目标检测算法[J]. 红外与激光工程, 2003, 32(5): 472-475.
[6]
[7]	Di Nan, Zhu Mingan. An algorithm for real-time tracking target in complex environment[J]. Journal of Image and Graphics, 2008, 13(9): 1758-1764. 邸男, 朱明. 一种复杂背景下的实时目标跟踪算法[J]. 中国图象图形学报, 2008, 13(9): 1758-1764.
[8]
[9]
[10]	Han Minghua, Yuan Naichang. A improved tracking Kalman filter using multilayer neural network[J]. Journal of Electronics Information Technology, 1998, 20(6): 739-744. 韩明华, 袁乃昌.多层神经网络在跟踪式卡尔曼滤波器中的应用[J]. 电子科学学刊, 1998, 20(6): 739-744.
[11]
[12]	Vaideh V, Chitra N, Chokkalingam M, et al. Neural network aided Kalman filtering for multitarget tracking applications[J]. Computers and Electrical Engineering, 2001, 27(2): 217-228.
[13]
[14]	Greg Welch, Gary Bishop. An Introduction to the Kalman Filter, TR95-041[R]. Chapel Hill: Department of Computer Science, University of North Carolina, 1995.
[15]
[16]	Li Shaojun, Zhu Zhenfu. Track-before-detect algorithm via Particle filtering[J]. Infrared and Laser Engineering, 2009, 38(2): 352-357. 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.
[17]	Md. Zahidul Islam, Chi-min Oh, Chil-Woo Lee. Real time moving object tracking by particle filter[C]//International Symposium on Computer Science and its Applications, CSA' 08, 2008: 347-352.
[18]
[19]	Fazlina Ahmat Ruslan, Ramli Adnan, Abd Manan Samad, et al. Parameters effect in sampling importance resampling (SIR) particle filter prediction and tracking of flood water level performance[C]//2012 12th International Conference on Control, Automation and Systems (ICCAS), 2012: 868-872.

Multi-target detection and tracking of video sequence based on Kalman_BP neural network

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Related

Proportional views