Joint detection and tracking of size-varying infrared targets based on block-wise sparse decomposition

The high variability of target size makes small target detection in Infrared Search and Track (IRST) a challenging task. A joint detection and tracking method based on block-wise sparse decomposition is proposed to address this problem. For detection, the infrared image is divided into overlapped blocks, and each block is weighted on the local image complexity and target existence probabilities. Target-background decomposition is solved by block-wise inexact augmented Lagrange multipliers. For tracking, label multi-Bernoulli (LMB) tracker tracks multiple targets taking the result of single-frame detection as input, and provides corresponding target existence probabilities for detection. Unlike fixed-size methods, the proposed method can accommodate size-varying targets, due to no special assumption for the size and shape of small targets. Because of exact decomposition, classical target measurements are extended and additional direction information is provided to improve tracking performance. The experimental results show that the proposed method can effectively suppress background clutters, detect and track size-varying targets in infrared images.     

基于运动检测的多车辆跟踪方法研究

针对交通监控场景中多目标粘连造成跟踪上的困难和前后两帧车辆关联困难,提出了区域运动相似性分割方法和相似度关联矩阵的解决方案;在运动目标检测过程中, 首先使用背景差分法提取运动区域,经过消除缺口、空洞和分离等处理,在运动区域所在范围内进行块匹配搜索和局部光流计算区域运动矢量,然后使用模糊聚类方法对运动矢量区域融合,完整的分割出粘连运动目标;在目标跟踪部分,目标跟踪建立在目标关联的基础上,提出建立连续两帧目标间距离和局部二元模式相似度关联矩阵的方法进行运动目标标定,从而实现多目标关联;使用公共视频库的图像序列进行测试,所提算法都能实现连续的跟踪和准确的运动目标分割,且处理速度快,表明了算法具有鲁棒性和适用性。     

Bearings-Only Target Motion Analysis Using a Dual-Waveband Infrared System

Infrared systems are widely used for target detection, designation and tracking. For example, an Infrared Search and Track (IRST) system, as a typical airborne or shipboard detecting device, is widely equipped for the remote target detection and tracking. In recent years, the problem of target motion analysis (TMA) and tracking has been studied increasingly extensively. In an airborne infrared system, the problem becomes more difficult due to absence of range information. In this paper, the infrared model and motion model of typical aerial targets are described. An airborne dual-waveband IRST system, which is quite familiar nowadays, is choosed for implementation of target motion analysis. Based on the above, a novel and more practical algorithm of target tracking via bearings-only measurements is formed and the major parameters are defined according to a typical airborne dual-waveband IRST system. Finally, data simulation is implemented, and the results demonstrate the new algorithm has a better performance than before for bearings-only target tracking.     

基于分块速度域改进迭代运动目标检测算法的红外弱小目标检测

提出一种基于分块速度域的迭代红外运动目标检测算法来解决传统算法计算量巨大这一难题.首先,采用二维最小均方差滤波器对红外序列图像进行滤波,获得包含弱小目标以及残差的红外序列图像.然后,通过在序列图像块的速度域上应用改进的迭代运动目标检测算法进行能量累积,从而将弱小目标的运动速度在速度域进行累积增强,达到检测弱小运动目标的目的.最后在解算出的速度值附近进行搜索,得到弱小目标运动的精确速度.利用此速度进行空域能量累积,得到叠加图像,在此图上进行目标检测.与传统方法相比较,几组实验结果显示,本文提出的方法大大缩短了检测的时间,而且本文方法的检测效果也较好.     

Detection method for small and dim targets from a time series of images observed by a space-based optical detection system

To revisit cataloged space targets, a space-based optical detection system normally observes space targets continuously in a target tracking mode. In the time series of images produced by continuous observation, there are not only the target but also complicated background clutter (a mass of stars) and noises. The existing method only can detect the target with an signal-to-noise ratio (SNR) greater than 6 from these images. This paper presents a detection method for the target with an SNR less than 6. The proposed method consists of an SNR enhancement algorithm and an adaptive background and noise suppression algorithm. Simulation and analytical results show the proposed method detects the target submerged in noise and background clutter when SNR is equal to 3 and the detection probability and the false alarm probability both reach very high performance. This proposed method can help solve the problem of revisiting some weak cataloged space targets.     

基于卷积特征选择的红外目标跟踪

对红外图像中的目标跟踪时,复杂的背景信息以及目标像素数较少等因素增加了红外目标跟踪难度,目标区域的图像块缺乏特征信息使得普通跟踪算法较易产生跟踪偏移问题。为解决此问题,提出了一种基于粒子滤波框架下的卷积特征选择的红外目标跟踪算法。首先,在初始目标块上提取少量图像块作为滤波器,进而获得表征能力更强的卷积特征。然后,采用在线提升算法对该特征进行选择,增加跟踪算法的精度和执行效率。最后,将贝叶斯分类器的响应作为粒子权值估计出目标状态。实验结果验证了所提算法的跟踪性能优于其他几种传统算法。     

一种基于特征提取的相关跟踪算法

介绍了一种能稳定快速跟踪复杂背景下目标的算法,该算法在传统相关跟踪算法的基础上进行改进.当目标进入红外(电视)摄像机视场时,视频信号中包含有目标信息和背景信息,信号处理器先将此信号进行数字化处理,形成具有一定灰度等级的数字化图像阵列,然后采用边缘检测、阈值分割等算法对包含有目标信息的图像进行边缘处理,提取出具有特征的目...     

联合多站阵元域数据的水下目标检测与跟踪

为了提高复杂海洋环境中目标的检测、跟踪性能,提出一种联合多站阵元域数据的水下目标检测与跟踪方法.该方法采用序列马尔科夫链蒙特卡洛思想对目标进行采样更新,通过对接收概率中的后验概率以及采样函数进行分解展开,并根据多站阵元域数据计算采样粒子的联合似然,在迭代过程中实现目标数目和目标状态的联合估计.研究结果表明,该方法对单目标的平均定位误差在较高信噪比下能够稳定在50 m以内,对多目标随机出入场景中新生及消失目标实现有效检测,同时对强干扰下弱目标及交叉目标实现有效检测跟踪。仿真结果和海试数据均验证该方法具有良好的目标检测与跟踪性能。      

Tensor Fukunaga-Koontz transform for small target detection in infrared images

Infrared small targets detection plays a crucial role in warning and tracking systems. Some novel methods based on pattern recognition technology catch much attention from researchers. However, those classic methods must reshape images into vectors with the high dimensionality. Moreover, vectorizing breaks the natural structure and correlations in the image data. Image representation based on tensor treats images as matrices and can hold the natural structure and correlation information. So tensor algorithms have better classification performance than vector algorithms. Fukunaga-Koontz transform is one of classification algorithms and it is a vector version method with the disadvantage of all vector algorithms. In this paper, we first extended the Fukunaga-Koontz transform into its tensor version, tensor Fukunaga-Koontz transform. Then we designed a method based on tensor Fukunaga-Koontz transform for detecting targets and used it to detect small targets in infrared images. The experimental results, comparison through signal-to-clutter, signal-to-clutter gain and background suppression factor, have validated the advantage of the target detection based on the tensor Fukunaga-Koontz transform over that based on the Fukunaga-Koontz transform.     

A tracking filter method of active sonar subject to unknown target maneuver

A tracking filter algorithm based on the maneuvering detection delay is presented in order to solve the fuzzy problem of target maneuver decision introduced by the measure?ment errors of active sonar. When the maneuvering detection is unclear, two target moving hypotheses, the uniform and the maneuver, derived from the method of multiple hypothesis tracking, are generated to delay the final decision time. Then the hypothesis test statistics is constructed by using the residual sequence. The active sonar?s tracking ability of unknown prior information targets is improved due to the modified sequential probability ratio test and the integration of the advantages of strong tracking filter and the Kalman filter. Simulation results show that the algorithm is able to not only track the uniform targets accurately, but also track the maneuvering targets steadily. The effectiveness of the algorithm for real underwater acoustic targets is further verified by the sea trial data processing results.     

Detection and tracking of targets in infrared images using Bayesian techniques

Simple yet robust techniques for detecting targets in infrared (IR) images are an important component of automatic target recognition (ATR) systems. In our previous works, we have developed IR target detection and tracking algorithms based on image correlation and intensity. In this paper, we discuss these algorithms, their performances and problems associated with them and then propose novel algorithms to alleviate these problems. Our proposed target detection and tracking algorithms are based on frequency domain correlation and Bayesian probabilistic techniques, respectively. The proposed algorithms are found to be suitable for real-time detection and tracking of static or moving targets, while accommodating for detrimental affects posed by the clutter and background noise. Finally, limitations of all these algorithms are discussed.     

单矢量水听器加窗直方图算法目标探测性能分析*

单个矢量水听器直方图算法具有良好的鲁棒性和目标方位估计性能,该文对直方图算法目标探测性能进行了分析和总结,并提出了一种基于目标方位估计的水中目标自主探测与跟踪算法,该算法可实现水中目标有无自主检测。仿真和消声水池测试结果表明,直方图算法实现目标自主跟踪所要求的信噪比需大于-7 dB,此时测向误差约为8°,-3 dB方位谱宽度在20°左右。海上试验数据分析表明,直方图算法对航速8.4 kn的水面航船在距离13.8 km范围内,可实现全程目标探测和跟踪,测向误差最优可达5?,在距离2 km时-3 dB方位谱宽度可达10°左右。     

目标机动未知下的有源声呐跟踪滤波方法

为解决由有源声呐测量误差而引入的目标机动判决模糊问题,提出了基于延迟机动检测的跟踪滤波算法。利用多假设跟踪方法,在目标机动检测模糊时,生成匀速及机动的两种目标运动假设以延迟最终的决策时间,基于残差序列构造假设检验统计量,实施序列似然比检验并融合强跟踪滤波器和Kalman滤波器的优点,提高了有源声呐对先验信息未知目标的跟踪能力。通过仿真分析表明,该算法不仅能够精确的跟踪匀速运动目标,而且能够稳定的跟踪机动目标。海试数据处理进一步验证了算法跟踪真实水声目标的有效性。      

液体可视图象序列中杂质小目标的检测方法（英）

在分析红外图象和可视图象差异的基础上,研究了一种适用于可视图象序列的运动小目标检测算法.受目标影响的象素点具有较弱的时间相关性,即相对普通象素点有较大的时间方差,根据此特征,用管道更新的方法产生新方差图象序列,再用管道更新的方法累积,增加信噪比,然后用基于统计均值的自适应门限方法分割出目标,最后将原图象的梯度倒数作为象素点强度的加权,有效地抑制灰度突变边界的传感器噪音,保存目标点.     

液体可视图象序列中杂质小目标的检测方法

在分析红外图象和视图象差异的基础上，研究了一种适用于可视图像序列的运动小目标检测算法。受目标影响的象素点具有较弱的时间相关性，即相对普通象素点有较大的时间方差，根据此特征，用管道更新的方法产生新方差图象序列，再用管道更新的方法累积，增加信噪比，然后用基于统计均值的自适应门限方法分割出目标，最后将原图象的梯度倒数作为象素点强度的加权，有效地抑制灰度突变边界的传感器噪音，保存目标点。     

复杂背景下目标检测的级联分类器算法研究

目标检测与跟踪一直是图像处理与计算机视觉领域的热门研究方向之一,其对军事上的成像制导、跟踪军事目标等以及民事方面的安防监控、智能人机交互等方面均有着重要的研究价值.将特征匹配问题看成是一种更普遍的二分类问题,将这种难解的高维计算变成二分类问题,使计算复杂度大大减小,这类方法以大数定律和贝叶斯法则为理论依据,本文提出一种非树形结构的分类器,并从理论上推导出其实现公式,将1bitBP特征应用到分类器中,同时采用计算量由小到大的三个分类器进行级联从而实现鲁棒精确的目标检测.从实验结果来看,本文算法能够对目标的尺度变化、旋转、部分遮挡、形变、模糊、背景变化等复杂情况有较好鲁棒性,并且检测精度相对较高,而本文算法的计算复杂度低、计算量小,有较高的应用价值.     

海天复杂背景下红外目标的检测跟踪算法

在分析海天复杂背景下红外目标图像特征的基础上,提出适合该环境的红外目标检测算法.该算法采用行均值相减的方法抑制海平面非线性温度场的影响,并进行中值滤波处理.对于更加复杂的环境,选用数学形态滤波法抑制背景中的大面积云团或海浪,从而确定出目标区域来进行目标图像的分割及增强.同时,综合使用图像捕获区域指定、运动目标检测法、弱目标的增强提取、记忆外推功能、数据融合加权跟踪方法,来保证在海天复杂背景下红外目标的可靠跟踪.实验表明,该算法能较好地处理海天复杂背景下红外目标的检测,且算法易于硬件实现,提高目标检测的实时效率.     

复杂背景灰度图像下的多特征融合运动目标跟踪

为解决低对比度、低信噪比、目标旋转、缩放等非理想状态给跟踪算法的研究带来的诸多困难,本文提出灰度图像多特征融合目标跟踪算法,保证在满足工程实践需要的条件下,能够对目标进行稳定的跟踪。算法首先对灰度图像利用Sobel算子求出梯度特征,将X、Y双方向的梯度特征与灰度特征相融合得到新特征,新特征在核密度函数下对低对比度,目标轮廓形状变化较大的情况有较高的适应性和稳定性,再利用背景建模的方法对提取的运动目标区域进行加权,降低非跟踪目标的权值,最后对融合后的加权特征目标利用改进MeanShift算法进行跟踪。通过大量的实验表明,该算法适应目标和背景的复杂变化,并且具有较强的鲁棒性,基本满足在复杂背景灰度图像下目标跟踪的工程实际需求。     

A target group tracking algorithm for wireless sensor networks using azimuthal angle of arrival information

<正>In this paper,we explore the technology of tracking a group of targets with correlated motions in a wireless sensor network.Since a group of targets moves collectively and is restricted within a limited region,it is not worth consuming scarce resources of sensors in computing the trajectory of each single target.Hence,in this paper,the problem is modeled as tracking a geographical continuous region covered by all targets.A tracking algorithm is proposed to estimate the region covered by the target group in each sampling period.Based on the locations of sensors and the azimuthal angle of arrival(AOA) information,the estimated region covering all the group members is obtained.Algorithm analysis provides the fundamental limits to the accuracy of localizing a target group.Simulation results show that the proposed algorithm is superior to the existing hull algorithm due to the reduction in estimation error,which is between 10%and 40%of the hull algorithm,with a similar density of sensors.And when the density of sensors increases,the localization accuracy of the proposed algorithm improves dramatically.     

基于FPGA的运动目标实时检测跟踪算法及其实现技术

运动目标检测跟踪有关的算法及其基于PC平台的实现已经比较成熟,但实时性较差。将采集的彩色视频流分成灰度和彩色两个数据流,灰度视频用于目标检测,彩色视频流用于跟踪显示。以经典的帧间差分法和背景差分法为基础,根据现场可编程门阵列(FPGA)的特点及片外同步动态存储器的存取控制要求,对这两个算法用FPGA逻辑单元进行了设计和实现。对原始彩色视频流和转换后的灰度视频流的存取使用乒乓操作,在滤波和形态学处理时使用了并行的流水线操作,极大地提高了算法的实时处理能力。在FPGA开发板上构建了一个彩色视频图像中运动目标检测跟踪系统,对系统性能进行了测试。实验结果表明,系统可在多种分辨率和帧率下进行运动目标进行实时检测跟踪;固定背景差分法对目标运动速度无限制,但当使用帧差法对快速运动目标进行有效的检测时,应使目标的帧差间距大于3.2像素。