Mixture Particle PHD Filter Based Multi-Target Visual Tracking
LIN Qing1,2,XU Xiao-Gang1,ZHAN Yong-Zhao1,LIAO Ding-An1,3,YANG Ya-Ping1
1.School of Computer Science and Communication Engineering,Jiangsu University,Zhenjiang 212013 2.School of Computer Science and Technology,Nanjing University of Science and Technology,Nanjing 210094 3.Changzhou Textile Garment Institute,Changzhou 213164
Abstract Aiming at the problem that particle filter is poor at consistently maintaining the multi-modality of the target distributions for multi-targets in a variable number of visual tracking,a multi-target visual tracking approach based on mixture particle probability hypothesis density (PHD) filter is proposed. The particles are clustered by the K-means algorithm,the classified particles are labeled and the particle filters are separately used for each classified particles. It improves the accuracy of target states estimation and effectively maintains the multi-modal distribution of the various objectives. The experimental results show that the proposed approach is an effective solution to the appearance,merger,separation and other multi-target tracking problems for the new target.
[1] Vermaak J,Doucet A,Pérez P. Maintaining Multi-Modality through Mixture Tracking // Proc of 9th IEEE International Conference on Computer Vision. Nice,France,2003,II: 1110-1116 [2] Ouyang Cheng,Ji Hongbing,Guo Zhiqiang. Improved Multiple Model Particle PHD and CPHD Filters. Acta Automatica Sinica,2012,38(3): 341-348 (in Chinese) (欧阳成,姬红兵,郭志强.改进的多模型粒子PHD和CPHD滤波算法.自动化学报,2012,38(3): 341-348) [3] Bar-Shalom Y,Fortmann T. Tracking and Data Association. San Diego,USA: Academic Press, 1988 [4] Reid D B. An Algorithm for Tracking Multiple Targets. IEEE Trans on Automatic Control,1979,24(6): 843-854 [5] Mahler R P S. Multitarget Bayes Filtering via First-Order Multitarget Moments. IEEE Trans on Aerospace and Electronic Systems,2003,39(4): 1152-1178 [6] Vo B N,Singh S,Doucet A. Sequential Monte Carlo Implementation of the PHD Filter for Multi-Target Tracking // Proc of the 6th International Conference on Information Fusion. Cairns,Australia,2003: 792-799 [7] Panta K,Vo B N,Singh S. Improved Probability Hypothesis Density (PHD) Filter for Multitarget Tracking // Proc of the 3rd IEEE International Conference on Intelligent Sensing and Information Processing. Bangalore,India,2005: 213-218 [8] King O,Forsyth D A. How Does Condensation Behave with a Finite Number of Samples // Proc of the 6th European Conference Computer Vision. Dublin,Ireland,2000,I: 695-709 [9] Mahler R. PHD Filters of Higher Order in Target Number. IEEE Trans on Aerospace and Electronic Systems,2007,43(4): 1523-1542 [10] Lin Qing,Xu Zhu,Wang Shitong,et al. Moving Objects Detection of Adaptive Gaussian Mixture Models on HSV. Computer Science,2010,37(10): 254-256 (in Chinese) (林 庆,徐 柱,王士同,等.HSV自适应混合高斯模型的运动目标检测.计算机科学,2010,37(10): 254-256) [11] Lin Qing,Shi Zhenjie,Zhan Yongzhao,et al. Tracking Regions of Interest in Moving People. Journal of Jiangsu University: Natural Science Edition,2010,31(2): 206-209 (in Chinese) (林 庆,史振杰,詹永照,等.基于感兴趣区域的运动人体跟踪.江苏大学学报:自然科学版,2010,31(2): 206-209) [12] Vo B N,Singh S,Doucet A. Sequential Monte Carlo Methods for Multi-Target Filtering with Random Finite Sets. IEEE Trans on Aerospace and Electronic Systems,2005,41(4): 1224-1245 [13] Lin Qing,Xu Jianghui,Wang Shitong,et al. Automatic Detection of Regions of Interest in Moving People. Journal of Computer-Aided Design Computer Graphics,2009,21(10):1446-1450 (in Chinese) (林 庆,徐剑晖,王士同,等.运动人体感兴趣区域的自动检测.计算机辅助设计与图形学学报,2009,21(10): 1446-1450 [14] Hu Shiqiang,Jing Zhongliang. Principles and Applications of Particle Filter. Beijing,China: Science Press,2010 (in Chinese) (胡士强,敬忠良,著.粒子滤波原理及其应用.北京:科学出版社,2010
2013, Vol. 26 
