Multi-target pig tracking algorithm based on joint probability data association and particle filter

In order to evaluate the health status of pigs in time,monitor accurately the disease dynamics of live pigs,and reduce the morbidity and mortality of pigs in the existing large-scale farming model,pig detection and tracking technology based on machine vision are used to monitor the behavior of pigs.However,it is challenging to efficiently detect and track pigs with noise caused by occlusion and interaction between targets.In view of the actual breeding conditions of pigs and the limitations of existing behavior monitoring technology of an individual pig,this study proposed a method that used color feature,target centroid and the minimum circumscribed rectangle length-width ratio as the features to build a multi-target tracking algorithm,which based on joint probability data association and particle filter.Experimental results show the proposed algorithm can quickly and accurately track pigs in the video,and it is able to cope with partial occlusions and recover the tracks after temporary loss.

A combination algorithm of Chaos optimization and genetic algorithm and its application in maneuvering multiple targets data association

The most important problem in targets tracking is data association which may be represented as a sort of constraint combinational optimization problem. Chaos optimization and adaptive genetic algorithm were used to deal with the problem of multi-targets data association separately. Based on the analysis of the limitation of chaos optimization and genetic algorithm, a new chaos genetic optimization combination algorithm was presented. This new algorithm first applied the "rough" search of chaos optimization to initialize the population of GA, then optimized the population by real-coded adaptive GA. In this way, GA can not only jump out of the "trap" of local optimal results easily but also increase the rate of convergence. And the new method can also avoid the complexity and time-consumed limitation of conventional way. The simulation results show that the combination algorithm can obtain higher correct association percent and the effect of association is obviously superior to chaos optimization or genetic algorithm separately. This method has better convergence property as well as time property than the conventional ones.

A new fast algorithm for multitarget tracking in dense clutter

A fast joint probabilistic data association （FJPDA） algorithm is proposed in tiffs paper. Cluster probability matrix is approximately calculated by a new method, whose elements βi^t（K） can be taken as evaluation functions. According to values of βi^t（K）, N events with larger joint probabilities can be searched out as the events with guiding joint probabilities, tiros, the number of searching nodes will be greatly reduced. As a result, this method effectively reduces the calculation load and nnkes it possible to be realized on real-thne, Theoretical ,analysis and Monte Carlo simulation results show that this method is efficient.

A RECURSIVE AND PARALLEL FAST JPDA ALGORITHM

Joint Probabilistic Data Association (JPDA) is a very fine optimal multitarget tracking and association algorithm in clutter. However, the calculation explosion effect in computation of association probabilities has been a difficulty. This paper will discuss a method based on layered searching construction of association hypothesis events. According to the method, the searching schedule of the association events between two layers can be recursive and with independence, so it can also be implemented in parallel structure. Comparative analysis of the method with relative methods in other references and corresponding computer simulation tests and results are also given in the paper.

Study on the Hungarian algorithm for the maximum likelihood data association problem

A specialized Hungarian algorithm was developed here for the maximum likelihood data association problem with two implementation versions due to presence of false alarms and missed detections. The maximum likelihood data association problem is formulated as a bipartite weighted matching problem. Its duality and the optimality conditions are given. The Hungarian algorithm with its computational steps, data structure and computational complexity is presented. The two implementation versions, Hungarian forest （HF） algorithm and Hungarian tree （HT） algorithm, and their combination with the naYve auction initialization are discussed. The computational results show that HT algorithm is slightly faster than HF algorithm and they are both superior to the classic Munkres algorithm.

Track Association for Dynamic Target Tracking System Based on AP Algorithm

Track association of multi-target has been recognized as one of the key technologies in distributed multiple-sensor data fusion system,and its accuracy directly impacts on the performance of the whole tracking system.A multi-sensor data association is proposed based on aftinity propagation(AP)algorithm.The proposed method needs an initial similarity,a distance between any two points,as a parameter,therefore,the similarity matrix is calculated by track position,velocity and azimuth of track data.The approach can automatically obtain the optimal classification of uncertain target based on clustering validity index.Furthermore,the same kind of data are fused based on the variance of measured data and the fusion result can be taken as a new measured data of the target.Finally,the measured data are classified to a certain target based on the nearest neighbor ideas and its characteristics,then filtering and target tracking are conducted.The experimental results show that the proposed method can effectively achieve multi-sensor and multi-target track association.

Study on multiple maneuvering targets tracking based on JPDA algorithm

A tracking algorithm for multiple-maneuvering targets based on joint probabilistic data association（JPDA）is proposed to improve the accuracy for tracking algorithm of traditional multiple maneuvering targets.The interconnection probability of the two targets is calculated,the weighted value is processed and the target tracks are obtained.The simulation results show that JPDA algorithm achieves higher tracking accuracy and provides a basis for more targets tracking.

ON THE EQUIVALENCE OF PDA ALGORITHM AND SIC-MMSE ALGORITHM

In this letter,by employing Gaussian distribution to approximate the probability density function(pdf) of the extrinsic information at the output of the multiuser detector as a function of the pdf of the input extrinsic messages,it is concluded that the Probabilistic Data Association(PDA) algorithm is equivalent to the Soft Interference Cancellation plus Minimum Mean Square Error algo-rithm(SIC-MMSE) .

A multi-target tracking algorithm based on Gaussian mixture model

Since the joint probabilistic data association(JPDA)algorithm results in calculation explosion with the increasing number of targets,a multi-target tracking algorithm based on Gaussian mixture model(GMM)clustering is proposed.The algorithm is used to cluster the measurements,and the association matrix between measurements and tracks is constructed by the posterior probability.Compared with the traditional data association algorithm,this algorithm has better tracking performance and less computational complexity.Simulation results demonstrate the effectiveness of the proposed algorithm.

遮挡环境下基于航海雷达的舰船目标跟踪方法研究

[目的]针对无人艇平台航海雷达在舰船目标跟踪中因障碍物遮挡而造成的目标航迹断裂问题,需综合利用环境先验信息,以提高雷达探测受限时目标跟踪航迹的连续性。[方法]通过提出遮挡环境下的综合概率数据互联(IPDA)算法,即O-IPDA,对环境遮挡情况进行实时预判;在遮挡环境中通过采用低检测概率和针对性设计的存在状态概率转移矩阵,以维持目标跟踪的连续性。[结果]在单目标跟踪场景中,当目标被暂时遮挡时,O-IPDA可以避免目标因持续性漏检而丢失,以保持跟踪航迹的稳定性,其中算法的抗遮挡能力取决于O-IPDA存在状态概率转移矩阵中的相关参数设置。[结论]O-IPDA目标跟踪方法具有一定的抗遮挡能力,可为航海雷达单目标跟踪研究提供参考。

基于动力学守恒定律的弹道目标关联方法

中段伴飞突防造成的各种有源或无源的弹道群目标会给雷达跟踪系统带来极大的挑战,导致其跟踪非本体实体目标或电假目标,从而出现关联错误的情况。中段实体弹道目标满足动力学守恒定律,可以充分利用该特性来改善跟踪系统的数据关联机制,因此提出一种基于动力学守恒定律的弹道目标概率数据关联(probability data association,PDA)方法,即在传统关联门筛选出有效量测的基础上,对动量矩和机械能进行联合统计检验,进一步剔除电假目标点迹或其他错误量测,并使用动量矩和机械能对加权关联概率进行修正。蒙特卡罗仿真验证了该方法的有效性。仿真结果表明,与传统PDA方法相比,所提方法能够有效抑制有源距离欺骗干扰和杂波的影响,提高跟踪精度。

基于路侧激光雷达的多目标检测与跟踪算法

为了检测与跟踪城市交叉口复杂环境下的道路目标,提出一种基于路侧激光雷达的多目标检测与跟踪算法。首先利用背景减除法滤除背景点云,随后融合5帧点云并利用曲率体素聚类算法检测目标得到3 D包围盒信息,之后通过自适应阈值的双门控和生存周期管理策略,有效提升关联精度并减少了目标丢失和误检,最后利用交互式多模型无迹卡尔曼滤波(IMM-UKF)和联合概率数据互联(JPDA)的融合算法完成道路目标的跟踪。试验结果表明,该算法在保证检测和跟踪性能基础上满足实时性要求,具有工程实用价值。

基于Hough变换的监视数据航迹关联算法研究

当前监视数据航迹关联算法在进行重叠宽度计算时,多以独立形式测定核算,效率较低,导致最终得出的平均偏差增大,为此提出对基于Hough变换的监视数据航迹关联算法的研究,根据当前计算需求,先进行航迹覆盖区域划分,采用多目标的方式,提升整体的计算效率,多目标形式计算出重叠关联宽度,以此为基础,构建Hough变换监视数据航迹关联测算模型,采用边缘航迹辅助修正实现测算。测试结果表明:针对选定的6个测试区域,经过连个周期的测定,最终在Hough变换的辅助下,所设计的监视数据航迹关联算法得出的航迹关联测算的平均偏差控制在1.2以下,说明此次所设计的监视数据航迹关联算法更为高效,针对性较强,具有实际的应用价值。

杂波环境下基于最大熵模糊聚类的JPDA算法

针对杂波环境下的多目标跟踪数据关联存在跟踪精度低、实时性差的问题,提出了一种基于最大熵模糊聚类的联合概率数据关联算法(joint probabilistic data association algorithm based on maximum entropy fuzzy clustering,MEFC-JPDA)。首先,采用最大熵模糊聚类求得的隶属度初步表征目标与有效量测之间的关联概率。其次,采用基于目标距离的量测修正因子对关联概率进行调整,并建立关联概率矩阵。最后,结合卡尔曼滤波算法,对目标的状态进行加权更新。仿真结果表明,所提算法在杂波环境下的跟踪性能相比现有的两种关联算法有较大提升,是一种有效的多目标跟踪数据关联算法。

一种基于激光雷达的水面多目标跟踪方法

文中提出概率数据互联与KM(Kuhn-Munkres)算法相结合的数据关联方法和基于生命周期理论的航迹管理方法.利用概率数据互联计算的匹配概率改进基于位置与形状特性的关联值计算方法,提高关联值的准确性.设计了一种基于生命周期理论的航迹管理方法,利用多阈值和多阶段的管理策略提高航迹的稳定性.结果表明:改进的关联方法能够提高存在虚警和航迹交叉情况下的跟踪正确率,而基于生命周期理论的航迹管理方法能够有效解决遮挡导致的航迹中断问题.

快速JPDA算法的递归和并行实现

联合概率数据关联算法（ＪｏｉｎｔＰｒｏｂａｂｉｌｉｓｔｉｃＤａｔａＡｓｏｃｉａｔｉｏｎ，ＪＰＤＡ）是密集杂波环境下一种良好的多目标数据关联跟踪算法。但是，当目标的数目增大时，关联概率计算时的计算量爆炸效应一直是一个难题。为降低计算量，有不少文献讨论了次优ＪＰＤＡ算法，但都是以降低关联跟踪性能为代价的。本文将从联合关联事件的构造出发，讨论关联假设事件的分层构造以达到降低计算量的目的。这里的层次可从０取值到某一Ｌ值，０层表示没有任何目标能够跟当前的观测数据关联。Ｌ层表示共有Ｌ个目标可以跟当前扫描得到的观测数据相关联。本文在关联事件的构造中，各层次的搜索具有递归性并可以独立进行，因而可以并行实现。文中还将本文的方法跟有关文献作了比较。

一种适于工程应用的多目标跟踪快速数据关联算法

提出了一种新的多目标跟踪快速数据关联算法 .重点分析了关联门相交区域中的公共回波对航迹更新的影响 ,并综合考虑了关联门内其余候选回波对航迹更新的作用 ,以很小的计算代价完成了后验概率的计算 .仿真表明 ,新算法以与PDAF算法接近的计算量 ,达到了接近于JPDAF算法的目标跟踪成功率 .

性能优化的跟踪门算法

本文提出一个基于数据关联性能评价的优化跟踪门算法 ,并通过它来减少跟踪门内来自非本目标的回波 ,最终达到提高多目标多传感器跟踪系统性能的目的 .与最优跟踪门相比 ,经理论分析和仿真数据表明 ,本算法有效改善了系统的性能 ,尤其在强干扰。

联合交互式多模型概率数据关联算法

通过对交互式多模型（ＩＭＭ）算法和概率数据关联（ＰＤＡ）算法的研究，指出Ｂａｒ－Ｓｈａｌｏｍ和Ｂｌｏｍ等提出的ＩＭＭＰＤＡ算法结构和理论上存在的问题。根据全局最优的思想，设计了一个最优波门，从而得到全局最优的量测集合，保证了系统的理论完整性和结构合理性。推导完成了新的联合ＩＭＭＰＤＡ算法——Ｃ－ＩＭＭＰ－ＤＡ算法。大量仿真计算验证了Ｃ－ＩＭＭＰＤＡ算法在减少计算量的同时，总体性能上较之原ＩＭＭＰＤＡ算法有大幅度提高。

广义概率数据关联算法

随着跟踪环境、跟踪对象和跟踪系统的不断变化、发展 ,目标与量测已很难仅仅以一一对应的关联关系来描述 ,这使得多目标跟踪中数据关联这一核心问题更具挑战 .JesusGarrcia、T .Kirubarajan和Bar Shalom等学者从智能方法或重复使用一对一分配JPDA等方面进行了研究 ,取得一定成效 ,但计算量和性能均未达到理想效果 .本文首先提出更符合实际情况的新的目标与量测相关联的可行性规则 ,给出广义联合事件的一种分割与组合方法 ,利用贝叶斯法则推导出了一种全局次优的广义概率数据关联算法 (GeneralizedProbabilityDataAssociation ,GPDA) .通过本文设计的各种典型环境的仿真计算表明 ,GPDA算法的性能在目标与量测无论是否在一一对应的情况下 ,全面优于JPDA算法 ,且由于新算法的设计技巧 ,使计算量和存储量也大大小于JPDA算法 ,为发展同时具有良好实时和关联性能的多目标跟踪算法给出了新的尝试 .