基于代价参考粒子滤波器组的多目标检测前跟踪算法  

A Multi-target Track-before-detect Algorithm Based on Cost-reference Particle Filter Bank

作　　者：卢锦 马令坤[1] 吕春玲 章为川 Sun Chang-Ming LU Jin;MA Ling-Kun;LV Chun-Ling;ZHANG Wei-Chuan;SUN Chang-Ming(School of Electrical Information and Artificial Intelligence,Shaanxi University of Science&Technology,Xi'an 710021,China;Schneider(Xi'an)Innovation&Technology Company Limited,Xi'an 710121,China;Institute of Integrated and Intelligent Systems,Griffith University,Brisbane 4111,Australia;Data61,Commonwealth Scientific and Industrial Research Organization,Sydney 1710,Australia)

机构地区：[1]陕西科技大学电子信息与人工智能学院,西安710021 [2]施耐德(西安)创新技术有限公司,西安710121 [3]格里菲斯大学集成与智能系统研究所,澳大利亚布里斯班4111 [4]联邦科学与工业研究组织Data61中心,澳大利亚悉尼1710

出　　处：《自动化学报》2024年第4期851-861,共11页Acta Automatica Sinica

基　　金：国家自然科学基金(61801281)资助。

摘　　要：针对图像序列中多目标检测和跟踪算法结构复杂、计算量大、性能降低等问题,提出一种基于代价参考粒子滤波器组的多目标检测前跟踪(Cost-reference particle filter bank based multi-target track-before-detect, CRPFB-MTBD)算法,将多目标跟踪问题转换为序贯地检测和跟踪多个单目标的问题.首先,采用代价参考粒子滤波器组序贯地估计所有可能单目标状态序列;其次,基于所有可能单目标状态序列的欧氏距离和累积代价确定目标数量;最后,根据累积代价判断每个目标出现和消失的具体时刻.仿真实验验证了CRPFB-MTBD的优良性能,与基于传统粒子滤波的多目标检测前跟踪算法(Particle filter based multi-target track-before-detect, PF-MTBD)、基于概率假设密度的检测前跟踪算法(Probability hypothesis density based track-before-detect, PHD-TBD)和基于伯努利滤波的检测前跟踪算法(Bernoulli based track-before-detect, Bernoulli-TBD)相比, CRPFB-MTBD的目标状态序列和数量估计结果最佳,且平均单次运行时间极短.Aiming at the problems of complex structure,increasing computation and decreasing performance of multiple targets detection and tracking algorithms in image sequences,a cost-reference particle filter bank based multi-target track-before-detect(CRPFB-MTBD)algorithm is proposed.In this work,the target tracking problem is converted into a problem of sequentially detecting and tracking multiple single targets.First,a cost reference particle filter bank is used to sequentially estimate all possible single targets’state sequences;secondly,the number of targets is determined based on the Euclidean distances and cumulative costs of all possible single targets’state sequences;finally,the specific moment when each target appears and disappears is determined based on the cumulative cost.The simulation experiment verified the excellent performance of CRPFB-MTBD.Compared with the traditional particle filter based multitarget track-before-detect(PF-MTBD)algorithm,probability hypothesis density based track-before-detect(PHD-TBD),and Bernoulli filter based track-before-detect(Bernoulli-TBD),CRPFB-MTBD has the best target state sequence and quantity estimation results,and the average single running time is extremely short.

关 键 词：多目标跟踪 检测前跟踪 粒子滤波 代价参考粒子滤波器组 滤波器组 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TN713[自动化与计算机技术—计算机科学与技术]