无人机多机动目标自主探测与跟踪

无人机机载相机图像中机动目标尺寸较小而且会发生显著变化,加上大量的背景噪声干扰,给目标探测和跟踪带来很大困难.针对这些问题,本文提出了一种在无人机机载相机图像序列中自主探测与跟踪多个机动目标的方法.首先,提取目标的图像数字特征并采用级联分类算法进行特征分类,得到目标的强分类器,对目标进行自主探测搜索.然后,基于全局最优关联算法对探测回波进行关联滤波,实现对多个机动目标的跟踪与识别,其中最优关联代价矩阵融合了距离和方向信息,提高了关联和跟踪的鲁棒性.将无人机航拍图像序列中的地面坦克作为目标进行实验,结果表明本文算法可以实现对多个机动目标的自主探测和跟踪,并具有较好的跟踪鲁棒性.     

分块检测尺度自适应的圆形目标相关跟踪

为解决传统基于核相关滤波器（KCF）的目标跟踪方法中跟踪框尺度无法自适应及目标被遮挡导致的跟踪失败问题,以空中加油视觉导航中圆形锥套目标的跟踪为例,提出了一种分块检测尺度自适应的圆形目标相关跟踪方法。利用目标外部边缘特征点的提取与轮廓拟合方法解决了圆形目标跟踪所存在的目标尺度变换问题,通过分块检测将跟踪过程中目标遮挡情况与跟踪失败情况区分开来,消除了遮挡情况下跟踪积累误差,实现了部分遮挡情况下目标的跟踪。实验结果表明,提出的圆形目标尺度自适应相关跟踪方法具有较高实时性、可靠性与准确性,实现了空中加油对接过程中对于锥套的稳定、可靠跟踪。     

基于改进TLD算法的无人机自主精准降落

四旋翼无人机（Unmanned Aerial Vehicle,UAV）在航拍、测绘、环境监测、快递等航空领域的广泛应用,对四旋翼无人机的可用性和可靠性提出了更高的要求,而其实现自主精准降落的功能是必不可少的。对目标进行快速鲁棒性跟踪是实现降落的重要基础,TLD（Tracking Learning Detector）算法为这一问题提供了一种有效的解决办法,虽然许多学者对其进行了研究并对传统的TLD算法进行了改进,但算法的跟踪精度及速度仍然难以满足无人机的降落要求。提出了一种基于TLD框架的目标跟踪算法来实现无人机与特定降落目标之间的相对定位。该算法在TLD框架下,提出一种基于目标形状特征自主确定降落目标的算法,提高了降落流程的自主性;用核相关滤波器（Kernelized Correlation Filter,KCF） 实现了TLD框架中的跟踪器,提高了算法的实时性、精准度及鲁棒性;同时在降落过程中采用一种基于方向梯度直方图特征（Histogram of Gradient,HOG）和支持向量机（Support Vector Machine,SVM） 的目标识别方法,以实现目标检测自矫正,保证长时间准确跟踪目标。在七类模拟无人机进行降落的视频集下验证了该算法,与其他三种跟踪算法进行对比,并进行实际降落测试。测试结果表明,该算法的鲁棒性和精准度均优于其他算法,处理速度可达到31.47?f/s,故而在TLD框架下采用核相关滤波器作为跟踪器,对跟踪及检测结果进行有效融合并提高算法实时性的同时,增加的检测自矫正环节保证了长时间跟踪的准确度,从而有效地实现了无人机全自主精准降落。     

Chance-constrained target tracking using sensors with bounded fan-shaped sensing regions

We present a robust target tracking algorithm for a mobile robot. It is assumed that a mobile robot carries a sensor with a fan-shaped field of view and finite sensing range. The goal of the proposed tracking algorithm is to minimize the probability of losing a target. If the distribution of the next position of a moving target is available as a Gaussian distribution from a motion prediction algorithm, the proposed algorithm can guarantee the tracking success probability. In addition, the proposed method minimizes the moving distance of the mobile robot based on the chosen bound on the tracking success probability. While the considered problem is a non-convex optimization problem, we derive a closed-form solution when the heading is fixed and develop a real-time algorithm for solving the considered target tracking problem. We also present a robust target tracking algorithm for aerial robots in 3D. The performance of the proposed method is evaluated extensively in simulation. The proposed algorithm has been successful applied in field experiments using Pioneer mobile robot with a Microsoft Kinect sensor for following a pedestrian.     

基于改进Kalman滤波的可见光极小目标TBD跟踪方法

针对无人机可见光图像极小目标跟踪问题,本文提出一种基于改进卡尔曼滤波的 (Tracking before detection,TBD)跟踪方法。首先利用检测算法定位目标位置作为卡尔曼滤波的测量值,检测过程中的匹配相似度参数作为卡尔曼滤波测量噪声协方差矩阵的参照依据,其次利用卡尔曼滤波建立跟踪框架预测下一帧的目标位置,最后检测模块以预测位置为 参考位置进行局部搜索,完成整个检测跟踪过程。为了提高跟踪效率,本文根据检测和预测位置积累误差判决检测模式,误差超过门限值则采取全局检测模式消除积累误差,否 则使用局部检测模式,降低TBD跟踪算法的运算复杂度。仿真实验证明,本文方法可以有效检测跟踪极小目标,提高跟踪的实时处理能力。     

基于多视角航拍配准的运动小目标检测与跟踪

针对存在3D场景遮挡的航拍视频运动小目标跟踪问题,提出一种基于多视角航拍配准的运动小目标检测和跟踪算法。该算法首先对图像序列间隔采样,利用Harris检测器提取全局特征点,通过Delaunay三角网对待配准图像实现初始匹配,然后利用整合变换模型计算差分图像,并利用累积能量检测出目标,最后通过卡尔曼运动滤波消除运动目标跟踪的抖动。实验结果表明,该算法对城市和郊区场景的航拍视频可以检测出最小30个像素的缓慢运动目标。     

Development of an algorithm for fast corner points detection

A method for detecting corner points in digital images is presented. The method is distinguished by high stability and efficiency compared with many method for detecting corner points developed earlier. The stability of corner detection is especially important in computer vision tasks connected with matching images of the same object, recovering digital surface models based on a set of images, and tracking objects. The overwhelming majority of algorithms detect equally well both correct corners and excessive points not corresponding to real corners of objects. The presented algorithm does have this disadvantage, and it can be used in frame-to-frame processing video in real time, e.g. in navigation systems of mobile robots and unmanned aerial vehicles. In addition, the proposed algorithm may be adapted to any data set since it is based on the machine learning method. The advantages of the developed method are demonstrated by an example of detection of corners in images of a typical hangar and in images with the international space station.     

面向嵌入式设备的高实时微小目标跟踪检测方法

传统主流目标检测算法在嵌入式平台无法兼具高实时性与高准确性,难以应用于边缘智能等领域。为解决微小目标跟踪检测在嵌入式平台实时应用的瓶颈,提出一种高实时微小目标跟踪检测方法。利用轻量化神经网络的骨干网络和路径聚合网络,对整体网络进行针对化的剪枝优化,同时深度融合相关滤波算法,提升针对微小目标跟踪检测的准确度和速度。在3D物体场景渲染器自建的军事微小目标数据集上的实验结果表明,在100像素的极小目标跟踪识别中,与DarkNet53-CSP方法相比,该方法检测精度大幅提高,在400~10 000像素的微小目标识别跟踪中,检测精度与检测速度优于DarkNet53和ResNeXt50+CSP等算法。     

基于多尺度建议框的目标跟踪误差修正方法

随着深度学习技术引入视觉目标跟踪领域,目标跟踪算法的精度和鲁棒性有了很大的提高。但在低空无人机跟踪目标的实际场景中,情况比较复杂,如相机的抖动、大量的遮挡、视角和焦距的改变等,使得跟踪算法的准确性受到极大挑战。目前的算法大多建立在目标外观变化缓慢的前提假设下,在跟踪的过程中不具备检测和修复漂移（跟踪误差）的能力。针对该问题,提出了一种基于多尺度建议框的目标跟踪误差修正方法。离线阶段,利用大量的已标注的目标样本训练基于多尺度建议框的目标跟踪修正模型,获取不同类别目标的先验知识。在线阶段在核相关滤波跟踪的基础上,依据相关响应置信度自适应评价的结果,通过目标跟踪修正模型不定期重新初始化目标的位置,避免了因为误差累积而导致跟踪失败。算法在无人机航拍数据集上进行了测试,结果表明,该跟踪算法在目标发生较大形变的情况下能较好的修正跟踪漂移问题。相比于其他几种算法,目标跟踪的成功率和精度分别提高了14.3%和3.1%。     

Conditions for target tracking with range-only information

This paper addresses the problem of guiding a mobile robot towards a target using only range sensors. The bearing information is not available. The target can be stationary or moving. It can be the source of some gas leakage or nuclear radiation or it can be some landmark or beacon or any manoeuvring vehicle. The mobile robot can be a ground vehicle or an aerial vehicle flying at a fixed altitude. In literature, many different strategies are proposed which use the range only measurement but they involve estimation of different parameters or have switching control strategy which make them difficult to implement. We propose two sets of conditions, one for stationary target and another for both stationary and moving target. Any control strategy, that will satisfy these conditions, can bring the robot arbitrarily close to the target. There are no restrictions on the initial conditions. Estimation of any parameter is not required. Some candidate controllers are presented that included continuous controllers and switching controllers. Simulations are carried out with these controllers to validate our result with and without measurement noise. Experimental results with ground mobile robot are presented.     

Performance evaluation of selective and adaptive heads clustering algorithms over wireless sensor networks

Target tracking in wireless sensor networks can be considered as a milestone of a wide range of applications to permanently report, through network sensors, the positions of a mobile target to the base station during its move across a certain path. While tracking a mobile target, a lot of open challenges arise and need to be investigated and maintained which mainly include energy efficiency and tracking accuracy. In this paper, we propose three algorithms for tracking a mobile target in wireless sensor network utilizing cluster-based architecture, namely adaptive head, static head, and selective static head. Our goal is to achieve a promising tracking accuracy and energy efficiency by choosing the candidate sensor nodes nearby the target to participate in the tracking process while preserving the others in sleep state. Through Matlab simulation, we investigate the performance of the proposed algorithms in terms of energy consumption, tracking error, sensor density, as well as target speed. The results show that the adaptive head is the most efficient algorithm in terms of energy consumption while static and selective static heads algorithms are preferred as far as the tracking error is concerned especially when the target moves rapidly. Furthermore, the effectiveness of our proposed algorithms is verified through comparing their results with those obtained from previous algorithms.     

考虑信息成功传递概率的多无人机协同目标最优观测与跟踪

针对考虑通信因素的多无人机协同目标最优观测与跟踪问题, 引入费舍信息矩阵对无人机探测所获取的信息进行表征, 考虑无线通信链路特性并对无人机间信息成功传递概率进行建模. 以无人机群体所获取的关于目标的信息量为指标函数, 分别建立是否考虑通信因素情况下的多机协同目标最优观测及跟踪问题模型. 对两种情况下的多机协同目标观测与跟踪进行仿真比较, 仿真结果验证了所建模型的有效性, 并体现了通信因素的重要影响.

     

基于面向对象Petri网的无人机协同追踪系统建模*

从离散事件系统角度出发,基于动态角色配置的思想,提出了协同追踪过程中无人机在搜索、协助者、领导者和空闲四个角色之间切换,动态构建团队追踪地面移动目标的协同追踪策略,采用面向对象Petri网对无人机、移动目标以及协同追踪策略组成的协同追踪系统进行建模研究。利用改进的人工势场法作为无人机的追踪算法,设计了一种协同追踪的情景对协同追踪系统进行了仿真,仿真结果表明了系统建模的可行性。     

面向航拍图像多运动目标的实时检测与识别

为了实现无人机(UAV)航拍图像中多运动目标的实时检测与识别,将静止目标和运动目标分别定义为背景和前景,利用图像稳化技术将航拍图像序列中的每帧与相邻帧对齐,克服UAV飞行动作对摄像机转动拍摄图像的影响;选取图像中的行人和车辆作为前景,分别使用哈尔(Haar-like)特征和级联分类器对图像中的目标进行检测和识别;利用密集光流计算两幅连续图像的运动矢量,从而区分静止目标(背景)和运动目标(前景),最终图像结果仅保留运动目标所在区域;将文章方法用于DroneVehicl航拍数据集实验,每秒平均帧数达到47.08 fps,检测精度为94%,并且表现出较高的召回率和F统计量,可达到实时检测与识别效果。     

Low-Cost Visual Tracking of a Landing Place and Hovering Flight Control with a Microcontroller

The growth of civil and military use has recently promoted the development of unmanned miniature aerial vehicles dedicated to surveillance tasks. These flying vehicles are often capable of carrying only a few dozen grammes of payload. To achieve autonomy for this kind of aircraft novel sensors are required, which need to cope with strictly limited onboard processing power. One of the key aspects in autonomous behaviour is target tracking. Our visual tracking approach differs from other methods by not using expensive cameras but a Wii remote camera, i.e. commodity consumer hardware. The system works without stationary sensors and all processing is done with an onboard microcontroller. The only assumptions are a good roll and pitch attitude estimation, provided by an inertial measurement unit and a stationary pattern of four infrared spots on the target or the landing spot. This paper details experiments for hovering above a landing place, but tracking a slowly moving target is also possible.     

Real-Time Feature Matching in Image Sequences for Non-Structured Environments. Applications to Vehicle Guidance

This paper describes an integrated vehicle control system with visual feedback. A general-purpose, low-level feature matching method, able to work in real time without any strict assumptions on the environment structure or camera parameters, generates low-level matching results, which are used as source of data for applications like mobile object tracking, among others. A generalized predictive path-tracking control approach keeps the vehicle on the trajectory defined by the moving target. In the low-level matching process, block-based features (windows) are selected and tracked along a stream of monocular images; least residual square error and similarity between clusters of features are used as constraints to select the right matching pair between multiple candidates. Real-time performance is achieved through optimized algorithms and a parallel DSP-based multiprocessor system implementation. Object detection and tracking is motion-based, and does not require a predefined model of the target. The integrated control system has been tested on the ROMEO-3R experimental vehicle.     

动态云台摄像机无人机检测与跟踪算法

为应对小型无人机的黑飞、滥飞对个人隐私、公共安全造成的威胁,本文采用高清云台摄像机定点巡航的方式对近地动态复杂背景中的无人机进行检测与跟踪,并提出了一种适用于动态云台摄像机的闭环无人机检测与跟踪算法,包含检测与跟踪两种模式。在检测模式下,本文设计了一种基于运动背景补偿的运动目标检测算法来提取分类候选区域,然后利用基于神经网络结构搜索得到的轻量级卷积神经网络对候选区域进行分类识别,可在不缩小高清视频图像的条件下实现无人机检测;在跟踪模式下,本文提出了一种结合卡尔曼滤波的局部搜索区域重定位策略改进了核相关滤波跟踪算法,使之在高清云台伺服追踪过程中仍能对目标进行快速稳定的跟踪;为将检测模式与跟踪模式结合在闭环框架中,本文还提出了一种基于检测概率和跟踪响应图状态的自适应检测与跟踪切换机制。实验表明,本文算法可应用于定点巡航状态的高清云台摄像机,实现近地复杂动态背景中无人机的实时准确检测、识别与快速跟踪。     

Flocking based distributed self-deployment algorithms in mobile sensor networks

In this paper, we address a novel deployment problem in isotropic mobile sensor networks. Sensors are to be relocated uniformly in a region of interest (ROI) centered at a target of interest (TOI) which could be stationary or mobile. With the assumption that relative direction of a sensor to the TOI can be recognized or inferred by devices equipped in the sensor, distributed control algorithms based on first-order and second-order dynamic models are proposed for both stationary and mobile TOI situations. The Lyapunov stabilities and coverage guarantee are provided. To further improve the deployment such as coverage holes inside the network and uniformity of the deployment, four assisted rules are also proposed. Then algorithms proposed for the situation of a stationary TOI are extended to anisotropic sensor networks. Simulations demonstrate the effective performances of the proposed algorithms.     

Moving people tracking with detection by latent semantic analysis for visual surveillance applications

The latent semantic analysis (LSA) has been widely used in the fields of computer vision and pattern recognition. Most of the existing works based on LSA focus on behavior recognition and motion classification. In the applications of visual surveillance, accurate tracking of the moving people in surveillance scenes, is regarded as one of the preliminary requirement for other tasks such as object recognition or segmentation. However, accurate tracking is extremely hard under challenging surveillance scenes where similarity among multiple objects or occlusion among multiple objects occurs. Usual temporal Markov chain based tracking algorithms suffer from the ‘tracking error accumulation problem’. The accumulated errors can finally make the tracking to drift from the target. To handle the problem of tracking drift, some authors have proposed the idea of using detection along with tracking as an effective solution. However, many of the critical issues still remain unsettled in these detection based tracking algorithms. In this paper, we propose a novel moving people tracking with detection based on (probabilistic) LSA. By employing a novel ‘twin-pipeline’ training framework to find the latent semantic topics of ‘moving people’, the proposed detection can effectively detect the interest points on moving people in different indoor and outdoor environments with camera motion. Since the detected interest points on different body parts can be used to locate the position of moving people more accurately, by combining the detection with incremental subspace learning based tracking, the proposed algorithms resolves the problem of tracking drift during each target appearance update process. In addition, due to the time independent processing mechanism of detection, the proposed method is also able to handle the error accumulation problem. The detection can calibrate the tracking errors during updating of each state of the tracking algorithm. Extensive, experiments on various surveillance environments using different benchmark datasets have proved the accuracy and robustness of the proposed tracking algorithm. Further, the experimental comparison results clearly show that the proposed tracking algorithm outperforms the well known tracking algorithms such as ISL, AMS and WSL algorithms. Furthermore, the speed performance of the proposed method is also satisfactory for realistic surveillance applications.     

面向无人机的视觉目标跟踪算法：综述与展望

近年来,无人机因其小巧灵活、智能自主等特点被广泛应用于民用和军事等领域中,特别是搜索侦察过程中首要的目标跟踪任务。无人机视觉目标跟踪场景的复杂性和运动目标的多变性,使得目标特征提取及模型建立困难,对目标跟踪性能带来巨大的挑战。本文首先介绍了无人机视觉目标跟踪的研究现状,梳理了经典和最新的目标跟踪算法,特别是基于相关滤波的跟踪算法和基于深度学习的跟踪算法,并对比了不同算法的优缺点。其次,归纳了常用的目标跟踪数据集和性能评价指标。最后,展望了无人机视觉目标跟踪算法的未来发展趋势。