Multiple Target Tracking Algorithm Based on Online Estimation of Target Birth Intensity
-
摘要: 针对多目标跟踪中未知的目标出生强度, 提出了基于Dirichlet分布的目标出生强度在线估计算法, 来改进概率假设密度滤波器在多目标跟踪中的性能. 算法采用有限混合模型来描述未知目标出生强度, 使用仅依赖于混合权重的负指数Dirichlet分布作为混合模型参数的先验分布. 利用拉格朗日乘子法推导了混合权重在极大后验意义下的在线估计公式; 混合权重在线估计过程利用了负指数Dirichlet分布的不稳定性, 驱使与目标出生数据不相关分量的消亡. 以随机近似过程为分量均值和方差的在线估计策略, 推导了基于缺失数据的分量均值与方差的在线估计公式. 在无法获得初始步出生目标先验分布的约束下, 提出了在混合模型上增加均匀分量的初始化方法. 以当前时刻的多目标状态估计值为出发点, 提出了利用概率假设密度滤波器消弱杂波影响的出生目标数据获取方法. 仿真结果表明, 提出的目标出生强度在线估计算法改进了概率假设密度滤波器在多目标跟踪中的性能.Abstract: As far as the unknown target birth intensity in multiple target tracking is concerned, an online estimation algorithm of target birth intensity is proposed to improve the performance of probability hypothesis density filter in multiple target tracking. The finite mixture model is adopted to model the unknown intensity of target birth. Dirichlet distribution with negative exponent parameters, which only depends on the mixing weights, is used as the prior distribution of parameters in the mixture model. The online estimation formulation of mixing weight is derived by Lagrange multiplier in the sense of maximum a posterior. The instability of Dirichlet distribution with negative exponent parameters is applied in driving the components irrelevant with birth targets to extinction during the online estimation procedure of mixing weights. Stochastic approximation procedure is regarded as the strategy of online estimation of component mean and covariance. The online estimation formulations of component mean and covariance are derived based on missing data. An initialization method is developed by adding a uniform distribution into the mixing model, under the constraint that no prior distribution of target birth is obtained in initialization. From the standpoint of current estimates of multiple target states, the method of achieving the data of birth targets, which makes full use of the ability of probability hypothesis density filter to reduce the effect of clutter, is presented. Simulation results show that the proposed online estimation algorithm of target birth intensity can improve the performance of probability hypothesis density filter in multiple target tracking.
-
[1] Pulford G E. Taxonomy of multiple target tracking methods. IEE Proceedings of Radar, Sonar and Navigation}, 2005, 152(5): 291-304[2] Blackman S, Popoli R. Design and Analysis of Modern Tracking Systems. Norwood: Artech House, 1999[3] Daley D, Vere-jones D. An Introduction to the Theory of Point Processes. Second Edition. New York: Springer-Verlag, 2002[4] Mahler R P S. Multitarget Bayes filtering via first-order multitarget moments. IEEE Transactions on Aerospace and Electronic Systems, 2003, 39(4): 1152-1178[5] Mahler R P S. Statistical Multisource-Multitarget Information Fusion. Norwood: Artech House, 2007[6] Erdinc O, Willett P, Bar-shalom Y. The bin-occupancy filter and its connection to the PHD filters. IEEE Transactions on Signal Processing, 2009, 57(11): 4232-4246[7] Vo B N, Singh S, Doucet A. Sequential Monte Carlo methods for multitarget filtering with random finite sets. IEEE Transactions on Aerospace and Electronic Systems, 2005, 41(4): 1224-1245[8] Whiteley N, Singh S, Godsill S. Auxiliary particle implementation of probability hypothesis density filter. IEEE Transactions on Aerospace and Electronic Systems, 2010, 46(3): 1437-1454[9] Liu W F, Han C Z, Lian F, Zhu H Y. Multitarget state extraction for the PHD filter using MCMC approach. IEEE Transactions on Aerospace and Electronic Systems, 2010, 46(2): 864-883[10] Vo B N, Ma W K. The Gaussian mixture probability hypothesis density filter. IEEE Transactions on Signal Processing, 2006, 54(11): 4091-4104[11] Pasha S A, Vo B N, Tuan H D, Ma W K. A Gaussian mixture PHD filter for jump Markov system models. IEEE Transactions on Aerospace and Electronic Systems, 2009, 45(3): 919-936[12] Clark D E, Bell J. Convergence results for the particle PHD filter. IEEE Transactions on Signal Processing, 2006, 54(7): 2652-2661[13] Clark D E, Vo B N. Convergence analysis of the Gaussian mixture PHD filter. IEEE Transactions on Signal Processing, 2007, 55(4): 1204-1212[14] Mahler R P S. PHD filters of higher order in target number. IEEE Transactions on Aerospace and Electronic Systems, 2007, 43(4): 1523-1543[15] Vo B T, Vo B N, Cantoni A. The cardinality balanced multi-target multi-Bernoulli filter and its implementations. IEEE Transactions on Signal Processing, 2009, 57(2): 409-423[16] Franken D, Schmidt M, Ulmke M. "Spooky action at a distance" in the cardinalized probability hypothesis density filter. IEEE Transactions on Aerospace and Electronic Systems, 2009, 45(4): 1657-1664[17] Vo B T, Vo B N, Cantoni A. Analytic implementations of the cardinalized probability hypothesis density filter. IEEE Transactions on Signal Processing, 2007, 55(7): 3553-3567[18] Zhang H J, Jing Z L, Hu S Q. Gaussian mixture CPHD filter with gating technique. Signal Processing, 2009, 89(8): 1521-1530[19] Punithakumar K, Kirubarajan T, Sinha A. Multiple-model probability hypothesis density filter for tracking maneuvering targets. IEEE Transactions on Aerospace and Electronic Systems, 2008, 44(1): 87-98[20] Lian Feng, Han Chong-Zhao, Liu Wei-Feng, Yuan Xiang-Hui. Multiple-model probability hypothesis density smoother. Acta Automatica Sinica, 2010, 36(7): 939-950 (连峰, 韩崇昭, 刘伟峰, 元向辉. 多模型概率假设密度平滑器. 自动化学报, 2010, 36(7): 939-950)[21] Clark D E, Bell J. Multi-target state estimation and track continuity for the particle PHD filter. IEEE Transactions on Aerospace and Electronic Systems, 2007, 43(4): 1441-1453[22] Panta K, Clark D, Vo B N. Data association and track management for the Gaussian mixture probability hypothesis density filter. IEEE Transactions on Aerospace and Electronic Systems, 2009, 45(3): 1003-1016[23] Vo B T, Vo B N, Cantoni A. Bayesian filtering with random finite set observations. IEEE Transactions on Signal Processing, 2008, 56(4): 1313-1326[24] Rezaeian M, Vo B N. Error bounds for joint detection and estimation of a single object with random finite set observation. IEEE Transactions on Signal Processing, 2010, 58(3): 1493-1506[25] Wang Y D, Wu J K, Kassim A A, Huang W M. Data-driven probability hypothesis density filter for visual tracking. IEEE Transactions on Circuits and Systems for Video Technology, 2008, 18(8): 1085-1095[26] Maggio E, Taj M, Cavallaro A. Efficient multitarget visual tracking using random finite sets. IEEE Transactions on Circuits and Systems for Video Technology, 2008, 18(8): 1016-1027[27] Maggio E, Cavallaro A. Learning scene context for multiple object tracking. IEEE Transactions on Image Processing, 2009, 18(8): 1873-1884[28] Clark D, Ruiz I T, Petilot Y, Bell J. Particle PHD filter multiple target tracking in sonar images. IEEE Transactions on Aerospace and Electronic Systems, 2007, 43(1): 409-416[29] Clark D, Ristic B, Vo B N, Vo B T. Bayesian multi-object filtering with amplitude feature likelihood for unknown object SNR. IEEE Transactions on Signal Processing, 2010, 58(1): 26-37[30] Zhang H J, Jing Z L, Hu S Q. Localization of multiple emitters based on the sequential PHD filter. Signal Processing, 2010, 90(1): 34-43[31] Mclachlan G, Peel D. Finite Mixture Models}. New York: John Wiley and Sons, 2000 \vskip 5mm[32] Davy M, Tourneret J Y. Generative supervised classification using Dirichlet process priors. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(10): 1781-1794[33] Figueiredo M A F, Jain A K. Unsupervised learning of finite mixture models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(3): 381-396[34] Hoffman J R, Mahler R P S. Multitarget miss distance via optimal assignment. IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans, 2004, 34(3): 327-336 }
点击查看大图
计量
- 文章访问数: 1997
- HTML全文浏览量: 63
- PDF下载量: 865
- 被引次数: 0