基于指标融合的跟踪算法性能评估度量标准
|
摘要
为了全面评估目标跟踪算法的性能,该文提出一种融合性能指标的新度量标准。在分析最优子模式分配度量标准基础上,新度量标准定义目标扩展状态包含跟踪算法的计算量信息,并在扩展状态的基本距离中引入了计算量距离概念,将计算量指标融入新度量标准的最优子模式分配距离中。新度量标准有效融合目标跟踪算法的跟踪精度、集合势误差和计算量指标,反映了跟踪算法的全面性能。单目标和多目标环境下的仿真实验结果验证了该标准的正确性和有效性。
In order to evaluate the performance of target tracking algorithms comprehensively,a new metric fusing performance indexes is proposed in this paper.Based on the analysis of optimal subpattern assignment(OSPA) metric,the new metric defines that the extended state of target includes the calculation amount information of tracking algorithms,and the concept of calculation amount distance is introduced into the base distance of extended state,the calculation amount index is incorporated into OSPA distance of the new metric.The new metric fuses the tracking precision,cardinality error and calculation amount index of the target tracking algorithms effectively and reflects the overall performance of the tracking algorithms.The simulation and experiment results under the environment of single target and multiple targets show the correctness and the validity of the proposed metric.
In order to evaluate the performance of target tracking algorithms comprehensively,a new metric fusing performance indexes is proposed in this paper.Based on the analysis of optimal subpattern assignment(OSPA) metric,the new metric defines that the extended state of target includes the calculation amount information of tracking algorithms,and the concept of calculation amount distance is introduced into the base distance of extended state,the calculation amount index is incorporated into OSPA distance of the new metric.The new metric fuses the tracking precision,cardinality error and calculation amount index of the target tracking algorithms effectively and reflects the overall performance of the tracking algorithms.The simulation and experiment results under the environment of single target and multiple targets show the correctness and the validity of the proposed metric.
引文
[1]RISTIC B,VO B N,CLARK D,et al.A metric for performance evaluation of multi-target tracking algorithms[J].IEEE Transactions on Signal Processing,2011,59(7):3452-3457.
[2]王芝,徐晓滨,刘伟峰,等.两类典型多目标跟踪算法的性能分析和比较[J].电子与信息学报,2010,32(7):1633-1637.WANG ZH,XU X B,LIU W F,et al.Performance analysis and comparison of two classic algorithms in multi-target tracking[J].Journal of Electronics and Information Technology,2010,32(7):1633-1637.
[3]杜思伟,林家俊.一种针对于状态估计算法的敏感指标构建方法[J].华东理工大学学报:自然科学版,2012,38(5):629-634.DU S W,LIN J J.An approach of constructing sensitive metric for state estimation algorithm[J].Journal of East China University of Science and Technology Natural Science Edition,2012,38(5):629-634.
[4]JAMSHAID A,CRISTINA D B,YEOL J C,et al.Some fixed point results for multi-valued mappings in partial metric spaces[J].Fixed Point Theory and Applications,2013,2013(1):1-13.
[5]HOFFMAN J R,MAHLER R P S.Multitarget miss distance via optimal assignment[J].IEEE Transactions on Systems,Man,Cybernetics-Part A:Systems and Humans,2004,34(3):327-336.
[6]SCHUHMACHER D,VO B T,Vo B N.A consistent me-tric for performance evaluation of multi-object filters[J].IEEE Transactions on Signal Processing,2008,56(8):3447-3457.
[7]HE X F.A track quality based metric for evaluating performance of multitarget filters[J].IEEE Transactions on Aerospace and Electronic Systems,2013,49(1):610-616.
[8]LIU Z X.A sequential GM-based PHD filter for a linear Gaussian system[J].Science China Information Science,2013,56(10):1-10.
[9]ZHANG S C,LI J X,WU L B,et al.A multiple maneuvering targets tracking algorithm based on a generalized pseudo-Bayesian estimator of first order[J].Journal of Zhejiang University SCIENCE C,2013,14(6):417-424.
[10]LUNDGREN M,SVENSSON L,HAMMARSSTRAND L.A CPHD filter for tracking with spawning models[J].IEEE Journal of Selected Topics in Signal Processing,2013,7(3):496-507.
[11]连峰,元向辉,陈辉.基于势概率假设密度滤波器的不可分辨目标跟踪算法[J].系统工程与电子技术,2013,35(12):2445-2451.LIAN F,YUAN X H,CHEN H.Tracking unresolved targets using cardinalized probability hypothesis density filter[J].Systems Engineering and Electronics,2013,35(12):2445-2451.
[12]文成林,吕冰,葛泉波.一种基于分布式滤波的数据融合算法[J].电子学报,2004,32(8):1264-1267.WEN CH L,LV B,GE Q B.A data fusion algorithm based on filtering step by step[J].Acta Elecctronica Sinica,2004,32(8):1264-1267.
[13]孟军英,刘教民,韩明,等.基于粒子滤波架构的实时目标跟踪算法研究[J].河北工业大学学报,2013,42(4):18-22.MENG J Y,LIU J M,HAN M,et al.Research of real-time target tracking based on particle filter framework[J].Journal of Hebei University of Technology,2013,42(4):18-22.
[14]蒋华锋,王瑶.基于K-S检验的直方图匹配目标跟踪算法[J].国外电子测量技术,2012,31(3):25-29.JIANG H F,WANG Y.Histogram matching target tracking algorithm based on K-S test[J].Foreign Electro-nic Measurement Technology,2012,31(3):25-29.
[15]万荣荣,李志宇,徐义华.基于卡尔曼滤波的高稳定度测温系统设计[J].电子测量技术,2012,35(4):29-31,41.WAN R R,LI ZH Y,XU Y H.Design of high-stability temperature measurement system based on Kalman[J].Electronic Measurement Technology,2012,35(4):29-31,41.
[16]庄晓燕,王厚军.基于卡尔曼滤波器的IEEE 1588时钟同步算法[J].电子测量与仪器学报,2012,26(9):747-751.ZHUANG X Y,WANG H J.IEEE 1588 clock synchroniz-ation algorithm based on Kalman filter[J].Journal of Electronic Measurement and Instrument,2012,26(9):747-751.
[17]王建平,徐恒,李奇越.基于卡尔曼滤波器的矿井移动节点定位算法研究[J].电子测量与仪表学报,2013,27(2):120-126.WANG J P,XU H,LI Q Y.Research of algorithm for mobile node localization in mine based on Kalman filtering[J].Journal of Electronic Measurement and Instrument,2013,27(2):120-126.
[18]王笑天,杨志家,王英男,等.双卡尔曼滤波算法在锂电池SOC估算中的应用[J].仪器仪表学报,2013,34(8):1732-1738.WANG X T,YANG J ZH,WANG Y N,et al.Application of dual extended Kalman filtering algorithm in the state-of-charge estimation of lithiumion battery[J].Chinese Journal of Scientific Instrument,2013,34(8):1732-1738.
[19]焉晓贞,罗清华.基于卡尔曼滤波的动态传感数据流估计方法[J].仪器仪表学报,2013,34(8):1847-1854.YAN X ZH,LUO Q H.Dynamic sensor data stream estimation method based on Kalman filtering[J].Chinese Journal of Scientific Instrument,2013,34(8):1847-1854.
[20]郑全普,雷刚,张凤伟,等.机载雷达精度评定方法研究[J].国外电子测量技术,2013,32(11):57-59,64.ZHENG Q P,LEI G,ZHANG F W,et al.Study on the methods of airborne radar accuracy evaluation[J].Foreign Electronic Measurement Technology,2013,32(11):57-59,64.
[21]徐梓皓,周召发,孙立江.基于卡尔曼滤波器的姿态角量测系统设计[J].电子测量技术,2014,37(1):1-4.XU Z H,ZHOU ZH F,SUN L J.Attitude measure system based on KF[J].Electronic Measurement Technology,2014,34(8):1-4.
[22]夏楠,邱天爽,李景春,等.一种卡尔曼滤波与粒子滤波相结合的非线性滤波算法[J].电子学报,2013,41(1):148-152.XIA N,QIU T S,LI J CH,et al.A nonlinear filtering algorithm combining the Kalman filter and the particle filter[J].Acta Elecctronica Sinica,2013,41(1):148-152.
[2]王芝,徐晓滨,刘伟峰,等.两类典型多目标跟踪算法的性能分析和比较[J].电子与信息学报,2010,32(7):1633-1637.WANG ZH,XU X B,LIU W F,et al.Performance analysis and comparison of two classic algorithms in multi-target tracking[J].Journal of Electronics and Information Technology,2010,32(7):1633-1637.
[3]杜思伟,林家俊.一种针对于状态估计算法的敏感指标构建方法[J].华东理工大学学报:自然科学版,2012,38(5):629-634.DU S W,LIN J J.An approach of constructing sensitive metric for state estimation algorithm[J].Journal of East China University of Science and Technology Natural Science Edition,2012,38(5):629-634.
[4]JAMSHAID A,CRISTINA D B,YEOL J C,et al.Some fixed point results for multi-valued mappings in partial metric spaces[J].Fixed Point Theory and Applications,2013,2013(1):1-13.
[5]HOFFMAN J R,MAHLER R P S.Multitarget miss distance via optimal assignment[J].IEEE Transactions on Systems,Man,Cybernetics-Part A:Systems and Humans,2004,34(3):327-336.
[6]SCHUHMACHER D,VO B T,Vo B N.A consistent me-tric for performance evaluation of multi-object filters[J].IEEE Transactions on Signal Processing,2008,56(8):3447-3457.
[7]HE X F.A track quality based metric for evaluating performance of multitarget filters[J].IEEE Transactions on Aerospace and Electronic Systems,2013,49(1):610-616.
[8]LIU Z X.A sequential GM-based PHD filter for a linear Gaussian system[J].Science China Information Science,2013,56(10):1-10.
[9]ZHANG S C,LI J X,WU L B,et al.A multiple maneuvering targets tracking algorithm based on a generalized pseudo-Bayesian estimator of first order[J].Journal of Zhejiang University SCIENCE C,2013,14(6):417-424.
[10]LUNDGREN M,SVENSSON L,HAMMARSSTRAND L.A CPHD filter for tracking with spawning models[J].IEEE Journal of Selected Topics in Signal Processing,2013,7(3):496-507.
[11]连峰,元向辉,陈辉.基于势概率假设密度滤波器的不可分辨目标跟踪算法[J].系统工程与电子技术,2013,35(12):2445-2451.LIAN F,YUAN X H,CHEN H.Tracking unresolved targets using cardinalized probability hypothesis density filter[J].Systems Engineering and Electronics,2013,35(12):2445-2451.
[12]文成林,吕冰,葛泉波.一种基于分布式滤波的数据融合算法[J].电子学报,2004,32(8):1264-1267.WEN CH L,LV B,GE Q B.A data fusion algorithm based on filtering step by step[J].Acta Elecctronica Sinica,2004,32(8):1264-1267.
[13]孟军英,刘教民,韩明,等.基于粒子滤波架构的实时目标跟踪算法研究[J].河北工业大学学报,2013,42(4):18-22.MENG J Y,LIU J M,HAN M,et al.Research of real-time target tracking based on particle filter framework[J].Journal of Hebei University of Technology,2013,42(4):18-22.
[14]蒋华锋,王瑶.基于K-S检验的直方图匹配目标跟踪算法[J].国外电子测量技术,2012,31(3):25-29.JIANG H F,WANG Y.Histogram matching target tracking algorithm based on K-S test[J].Foreign Electro-nic Measurement Technology,2012,31(3):25-29.
[15]万荣荣,李志宇,徐义华.基于卡尔曼滤波的高稳定度测温系统设计[J].电子测量技术,2012,35(4):29-31,41.WAN R R,LI ZH Y,XU Y H.Design of high-stability temperature measurement system based on Kalman[J].Electronic Measurement Technology,2012,35(4):29-31,41.
[16]庄晓燕,王厚军.基于卡尔曼滤波器的IEEE 1588时钟同步算法[J].电子测量与仪器学报,2012,26(9):747-751.ZHUANG X Y,WANG H J.IEEE 1588 clock synchroniz-ation algorithm based on Kalman filter[J].Journal of Electronic Measurement and Instrument,2012,26(9):747-751.
[17]王建平,徐恒,李奇越.基于卡尔曼滤波器的矿井移动节点定位算法研究[J].电子测量与仪表学报,2013,27(2):120-126.WANG J P,XU H,LI Q Y.Research of algorithm for mobile node localization in mine based on Kalman filtering[J].Journal of Electronic Measurement and Instrument,2013,27(2):120-126.
[18]王笑天,杨志家,王英男,等.双卡尔曼滤波算法在锂电池SOC估算中的应用[J].仪器仪表学报,2013,34(8):1732-1738.WANG X T,YANG J ZH,WANG Y N,et al.Application of dual extended Kalman filtering algorithm in the state-of-charge estimation of lithiumion battery[J].Chinese Journal of Scientific Instrument,2013,34(8):1732-1738.
[19]焉晓贞,罗清华.基于卡尔曼滤波的动态传感数据流估计方法[J].仪器仪表学报,2013,34(8):1847-1854.YAN X ZH,LUO Q H.Dynamic sensor data stream estimation method based on Kalman filtering[J].Chinese Journal of Scientific Instrument,2013,34(8):1847-1854.
[20]郑全普,雷刚,张凤伟,等.机载雷达精度评定方法研究[J].国外电子测量技术,2013,32(11):57-59,64.ZHENG Q P,LEI G,ZHANG F W,et al.Study on the methods of airborne radar accuracy evaluation[J].Foreign Electronic Measurement Technology,2013,32(11):57-59,64.
[21]徐梓皓,周召发,孙立江.基于卡尔曼滤波器的姿态角量测系统设计[J].电子测量技术,2014,37(1):1-4.XU Z H,ZHOU ZH F,SUN L J.Attitude measure system based on KF[J].Electronic Measurement Technology,2014,34(8):1-4.
[22]夏楠,邱天爽,李景春,等.一种卡尔曼滤波与粒子滤波相结合的非线性滤波算法[J].电子学报,2013,41(1):148-152.XIA N,QIU T S,LI J CH,et al.A nonlinear filtering algorithm combining the Kalman filter and the particle filter[J].Acta Elecctronica Sinica,2013,41(1):148-152.