Multiple target tracking using constrained MAP data association

A new scheme is introduced for data association for multiple target tracking. The scheme is formulated using the MAP estimation method and a new energy function. The natural constraints between targets and measurements are reflected in the energy function     

Evolutionary algorithm for data association and IMM-based target tracking in IR image sequences

Simultaneous tracking of multiple maneuvering and non-maneuvering targets in the presence of dense clutter and in the absence of any a priori information about target dynamics is a challenging problem. A successful solution to this problem is to assign an observation to track for state update known as data association. In this paper, we have investigated tracking algorithms based on interacting multiple model to track an arbitrary trajectory in the presence of dense clutter. The novelty of the proposed tracking algorithms is the use of genetic algorithm for data association, i.e., observation to track fusion. For data association, we examined two novel approaches: (i) first approach was based on nearest neighbor approach and (ii) second approach used all observations to update target state by calculating the assignment weights for each validated observation and for a given target. Munkres’ optimal data association, most widely used algorithm, is based on nearest neighbor approach. First approach provides an alternative to Munkres’ optimal data association method with much reduced computational complexity while second one overcomes the uncertainty about an observation’s source. Extensive simulation results demonstrate the effectiveness of the proposed approaches for real-time tracking in infrared image sequences.     

杂波环境下的粒子滤波器数据关联方法

为了解决杂波环境下多机动目标的数据关联难题,提出了一种将粒子滤波器(PF)和联合概率数据关联(JPDA)相结合的数据关联算法,该方法首先应用粒子滤波方法对目标的状态进行采样,得到样本(粒子),并结合量测,通过JPDA方法计算得到联合互连事件的关联概率,而该关联概率实际上就是PF中粒子的权值。通过选取适当的有效采样尺度作为衡量PF退化现象的测度,采用重要性重采样技术克服了标准PF的退化现象,降低了算法的计算量。仿真结果表明,粒子滤波方法可以较好地解决杂波环境下跟踪多机动目标的数据关联问题;重要性重采样PF的计算复杂度低于标准PF。     

基于空间直方图的海杂波背景下红外目标跟踪

空间直方图融合了目标的灰度分布信息和灰度的空间分布信息,比传统的灰度直方图更具有目标鉴别能力。为实现海杂波背景下稳健跟踪红外目标,本文在基于粒子滤波算法的红外目标跟踪系统框架中,将加权样本集表示红外目标的状态后验概率分布;采用简单的随机漂移模型表示系统状态模型;利用目标区域的空间直方图描述红外目标,其中通过核概率密度估计建立红外目标的灰度分布,然后统计灰度分布的空间信息建立空间直方图;通过空间直方图的相似度定义来建立系统观测概率模型,最终提出一种在海杂波背景下的基于空间直方图的粒子滤波红外目标跟踪算法。实     

Probabilistic data association for symbol detection in MIMO systems

The application of the probabilistic data association (PDA) algorithm to the symbol detection in narrowband spatial multiplexed multiple input multiple output systems with known channel state information is investigated. The performance of a new proposed complex formulation of PDA algorithm, which matches the full parameters of the complex Gaussian distribution (which are the mean vector, covariance matrix and the pseudo-covariance matrix), is compared with the real vector formulation of a generalised PDA algorithm and a PDA implementation, which employs complex Gaussian approximations with matched mean and covariance only     

A comparison study on node clustering techniques used in target tracking WSNs for efficient data aggregation

Wireless sensor applications are susceptible to energy constraints. Most of the energy is consumed in communication between wireless nodes. Clustering and data aggregation are the two widely used strategies for reducing energy usage and increasing the lifetime of wireless sensor networks. In target tracking applications, large amount of redundant data is produced regularly. Hence, deployment of effective data aggregation schemes is vital to eliminate data redundancy. This work aims to conduct a comparative study of various research approaches that employ clustering techniques for efficiently aggregating data in target tracking applications as selection of an appropriate clustering algorithm may reflect positive results in the data aggregation process. In this paper, we have highlighted the gains of the existing schemes for node clustering‐based data aggregation along with a detailed discussion on their advantages and issues that may degrade the performance. Also, the boundary issues in each type of clustering technique have been analyzed. Simulation results reveal that the efficacy and validity of these clustering‐based data aggregation algorithms are limited to specific sensing situations only, while failing to exhibit adaptive behavior in various other environmental conditions. Copyright © 2016 John Wiley & Sons, Ltd.     

MONO－PULSE RADAR 3－D IMAGING TECHNIQUES FOR TARGET IN STEPPED TRACKING MODE

A method for mono-pulse radar 3-D imaging in stepped tracking mode is presented and the amplitude linear modulation of error signals in stepped tracking mode is analyzed with its compensation method followes,so the problem of precisely tracking of target is solved.Finally the validity of these methods is proven by the simulation results.     

基于蚁群——遗传算法的改进多目标数据关联方法

将蚁群算法与遗传算法相结合,提出一种快速实现多目标数据关联的AC-GADA(ant colony-genetic algo-rithm data association)算法,该算法利用种群个体携带信息素,并改进了全局信息素扩散模型,同时为了提高算法的收敛速度并且避免局部极值的出现,引入了交叉变异策略和种群适应度模型,通过大量的实验数据证明,该算法在获得较高关联准确率的同时可以有效地提高关联速度。     

Optimum antenna polarizations for target discrimination in clutter

The concept of Stokes vectors and Stokes target operators is used to obtain optimum antenna polarizations for target discrimination in the presence of background clutter. The analysis involves a constrained maximization of the ratio of two bilinear forms, representing the signal-to-interference power ratio.     

Integrated real-time estimation of clutter density for tracking

The spatial density of false measurements is known as clutter density in signal and data processing of targets. It is unknown in practice and its knowledge has a significant impact on the effective processing of target information. This paper presents in the first time a number of theoretically solid estimators for clutter density based on conditional mean, maximum likelihood, and method of moments, respectively. They are computationally highly efficient and require no knowledge of the probability distribution of the clutter density. They can be readily incorporated into a variety of trackers for performance improvement. Simulation verification of the superiority of the proposed estimators to the previously used heuristic ones is also provided     

多目标跟踪中基于特征辅助的概率数据关联算法

在传统的多目标跟踪系统中,数据关联仅利用了那些与目标状态向量直接相关的信息。在此提出了一种基于广义概率数据关联(GPDA)的新的关联算法即特征辅助跟踪(FAT)算法。该算法同时利用了目标的特征信息和状态信息进行数据关联,较好地解决了在密集杂波环境下对近目标的跟踪问题。最后以目标的一维距离像信息为例进行仿真,仿真结果表明,所提出的算法使跟踪性能优于传统的概率数据关联。     

采用目标假设的三被动式传感器数据关联算法

被动式传感器目标跟踪是多传感器多目标跟踪领域的一个重要研究方向。针对三被动式传感器多目标定位系统全局最优数据关联的三维分配问题,在允许传感器漏检和虚警的情况下,通过分析拉格朗日松弛算法,提出一种假定真实目标的快速收敛算法。该算法通过粗关联假定真实目标并重新修改代价矩阵,然后进行细关联,使得拉格朗日松弛算法在保证关联精度的前提下有效地提高了收敛速度。理论分析和实验结果表明,该算法提高了数据关联的速度,并在一定程度上提高了关联准确率。     

基于粒子滤波和数据关联的多目标跟踪算法

针对图像序列中低信噪比条件下的点状多目标跟踪问题,在已获得各个目标初始信息的基础上,结合粒子滤波和联合概率数据关联(JPDA),研究了一种基于单帧检测的实时多目标跟踪算法.介绍了其基本思想和具体算法实现步骤,并在MATLAB仿真环境下实现了该跟踪算法.实验仿真结果表明,该算法能够准确跟踪作任意轨迹运动的多个点状动目标,具有良好的实时性与准确性.     

复合高斯杂波中自适应目标检测算法

本文研究了复合高斯杂波中已知给定距离-多普勒单元多普勒频率、未知目标幅度的自适应检测目标的问题。杂波散斑分量用未知自回归系数的自回归过程建模表示,通过求解自回归系数和目标幅度的最大似然估计,推导得到复合高斯杂波中的广义似然比检测器。在目标信号向量维数很大的情况下,该检测器对激励信号方差与纹理分量的乘积具有恒虚警率特性。数值仿真表明该检测器性能优于正则化自适应匹配滤波器。最后分析了目标信号向量维数、自回归模型阶数在不同信杂比下对检测器性能的影响。      

基于实测数据的宽带雷达杂波模型分析与运动目标检测

给出了对一组实测宽带雷达回波的分析和处理,该宽带雷达回波数据是利用一部X波段1 GHz带宽的微型宽带雷达对城市中的某处立交桥观测得到的.分析了杂波幅度的分布和杂波的功率谱密度,得出了幂次函数形式的杂波经验谱.利用子带多普勒处理和Chirp-Z变换的方法对回波数据进行了杂波抑制,动目标回波多普勒色散补偿和相参积累处理.结果表明:该方法使杂波功率显著降低,而目标回波得到了增强.     

Probabilistic Kalman filter for moving object tracking

Kalman filter has been successfully applied to tracking moving objects in real-time situations. However, the filter cannot take into account the existing prior knowledge to improve its predictions. In the moving object tracking, the trajectories of multiple targets in the same environment could be available, which can be viewed as the prior knowledge for the tracking procedure. This paper presents the probabilistic Kalman filter (PKF) that is able to take into account the stored trajectories to improve tracking estimation. The PKF has an extra stage after two steps of the Kalman filter to refine the estimated position of the targets. The refinement is obtained by applying the Viterbi algorithm to a probabilistic graph, that is constructed based on the observed trajectories. The graph is built in the offline situation and could be adapted in the online tracking. The proposed tracker has higher accuracy compared to the standard Kalman filter and could handle widespread problems such as occlusion. Another significant achievement of the proposed tracker is to track an object with anomalous behaviors by drawing an inference based on the constructed probabilistic graph. The PKF was applied to several manually-built videos and several other video-bases containing severe occlusions, which demonstrates a significant performance in comparison with other state-of-the-art trackers.     

Multifractal correlation characteristic for radar detecting low-observable target in sea clutter

This paper studies the multifractal correlation characteristic of sea clutter which can be applied to radar low-observable target detection in sea clutter. Multifractal correlation is a generalization of multifractal ‘single point’ statistic, and takes the spatial correlation of two points with different singularity intensity into account. The important content of this paper is the derivation of multifractal correlation from multifractal, and obtains the multifractal correlation spectrum. Then the multifractal correlation spectrum is used as a characteristic to analyze the membership degree to match board. In fact, radar target detection can be regarded as a binary-classification question, therefore the SVM (support vector machine) is applied to target detection. In the end, the real sea clutter under several conditions is used to verify the method proposed in this paper. The results show that the method has a strong performance of target detection and sea clutter suppression.     

Nonstationary probabilistic target and clutter scattering models

A probabilistic model for nonstationary and/or nonhomogeneous clutter and target scattering is proposed and developed. The first-order probability density of the scattered power is treated as the expected value of a conditional density that is a function of random parameters. The family of gamma densities is a general solution for the density function of the intensity reflected by objects comprised of several scatterers and is selected as the conditional density. In the general case, the gamma density is a function of two parameters: the mean and the inverse of the normalized variance. Assuming various distributions for a random mean, expressions for the first-order density of the scattered power are derived and used to explain previous experimental and theoretical results. An example of detection performance for nonstationary target fluctuation based on the developed model is also presented.     

一种相控阵雷达自适应数据率目标跟踪算法

为了提高相控阵雷达机动目标跟踪的系统资源利用率,给出了一种高效快速的自适应数据率交互多模型目标跟踪算法（FAIMM）,将目标数据更新时间与模型概率关联起来,使每一节拍的数据更新率随目标的瞬时机动特性自适应调整。仿真结果表明,与AIMM跟踪算法相比,自适应数据率跟踪算法准确率更高,系统资源消耗更少。