多特征重检测的相关滤波无人机视觉跟踪

无人机视觉跟踪是视觉跟踪未来应用的核心领域,其由于跟踪目标像幅较小、表 观不清且易受到无人机飞行姿态多变、飞行稳定性差等因素的影响而难以对目标进行鲁棒的跟 踪,特别是发生跟踪遮挡时,算法跟踪漂移后无法进行模型的更新。为提高无人机视频的跟踪 效果,提出一种多特征重检测跟踪方法。首先采用多特征融合的方式提高跟踪算法在无人机跟 踪特征的判别性。其次目标在出现遮挡时,扩大搜索区域,采用滑动窗口采样找到置信度最高 的目标区域并实现模型更新。通过一系列无人机视频实验结果表明,该算法在遇到遮挡问题时 具有较好的鲁棒性,能够提高无人机在目标跟踪过程中的准确性。     

四旋翼无人机自主移动降落方法研究

近年来,四旋翼无人机在自主完成各种复杂任务中扮演着愈发重要的角色。移动降落技术是无人机智能处理系统的关键技术,包含3个环节:目标检测、目标跟踪、位置预估及降落。提出了一种基于Apriltags的跟踪降落算法,提升了识别性能,通过Kalman滤波及拟合函数等方法预估运动目标轨迹,采用PID算法控制无人机稳定飞行、快速响应,实现了无人机对移动目标的智能识别、稳定跟踪及移动降落。     

一种基于无人机云台的视觉目标跟踪方法

鉴于无人机视觉目标跟踪时会遇到目标遮挡、目标尺度变化等问题,同时目标跟踪方法复杂度受到云台芯片算力的严格限制,提出了一种基于无人机云台的视觉目标跟踪方法。基于多个特征相关滤波器的自适应权重融合来提升目标位置预测的精度。引入长宽相互独立变化的尺度变化池策略,解决无人机目标跟踪过程中目标尺度估计问题。通过设置模板检测和轨迹预测模块来有效地对目标跟踪过程中的遮挡状态进行检测和处理。该方法可在无人机云台芯片中实现实时跟踪,且在公开无人机目标跟踪数据集和自采集数据集中取得了很好的跟踪效果,与基线方法相比,成功率提升了10.7个百分点,准确率提升了3个百分点。     

基于无人机跟踪的目标反遮挡算法

随着计算机视觉和无人机的蓬勃发展,目标跟踪是当下研究热点之一;但对于该方向,目前存在大量问题,目标的遮挡就是其中之一;因此,为了解决该问题,在此提出了一种无人机抗遮挡目标跟踪算法;该算法基于均值漂移目标跟踪算法;采用动态参数选择策略和双预测机制相结合的方法来解决目标跟踪过程中的目标遮挡问题;为了避免目标丢失,算法中采用目标检测的方法;大量实验表明,在与现有的算法比较过程中,该算法展示出了对遮挡目标跟踪有较好的鲁棒性和实时性;与其他算法相比,其精度提高约了10%,平均目标符合率提高了约15%。     

融合视觉显著性再检测的孪生网络无人机目标跟踪算法

针对旋翼飞行器在跟踪过程中目标尺度变化、快速运动、视角变化等问题,本文提出了一种基于MobileNetV2的孪生网络目标跟踪算法,可在无人机机载处理器上实时运行。该算法主要包含目标得分估计模块与目标尺度估计模块两个部分。结合多特征融合的策略,可准确预测出目标位置与目标框IoU,同时以目标框IoU为指导,利用梯度上升法对目标框进行迭代修正,进一步提升预测精度。针对完全遮挡而导致的目标跟丢问题,本文设计了一个基于视觉显著性的目标再检测算法,该算法可实时高效地预测出图像的显著性区域,以指导对目标的再检测,进而恢复跟踪。最后,通过标准无人机跟踪数据集测试与实际无人机跟踪实验,验证了算法的可行性。     

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%。     

基于改进PP-YOLO和Deep-SORT的多无人机实时跟踪算法

无人机（UAV）目标尺寸较小,多架无人机之间特征也不明显,且鸟类和飞虫的干扰给无人机目标的准确检测和稳定跟踪带来了巨大挑战。针对传统目标检测算法对小目标无人机检测性能差、跟踪不稳定的问题,提出一种基于改进PP-YOLO和Deep-SORT的多无人机实时跟踪算法。首先,将压缩-激励模块融入PP-YOLO检测算法中,以实现对无人机目标的特征提取和检测;其次,在ResNet50-vd结构中引入Mish激活函数,以解决反向传播过程中的梯度消失问题,并进一步提升检测精度;然后,采用Deep-SORT算法来实时跟踪无人机目标,并将提取外观特征的主干网络更换为ResNet50,从而改善原有网络对微小外观感知能力弱的状况;最后,引入损失函数Margin Loss,既提高了类别可分性,又加强了类内紧度和类间差异。实验结果表明,所提算法的检测平均精度均值（mAP）相比原始PP-YOLO算法提升了2.27个百分点,跟踪准确性相对于原始Deep-SORT算法提升了4.5个百分点。所提算法的跟踪准确性可达91.6%,能够实时跟踪600 m以内多架无人机目标,有效解决了跟踪过程中的“丢帧”问题。     

多特征融合的次峰抑制无人机目标跟踪

无人机目标跟踪是计算机视觉一个热门的研究方向。多特征融合后的响应通常含有噪声，为了解决这个问题，提出一种基于多特征融合的次峰响应抑制的无人机目标跟踪算法。提取HOG特征和二维颜色属性特征，并进行融合产生响应。对次峰响应进行抑制以去除噪声，将多个次峰聚合为一个中心主峰。引入自适应模型更新策略进一步增强算法的鲁棒性。在UAV123和VisDrone2019数据集上进行实验，结果显示该算法在无人机的快速运动、视角变化等挑战场景中表现出较好的跟踪精度和鲁棒性。     

基于深度学习的多旋翼无人机单目视觉目标定位追踪方法

针对无人机对目标的识别定位与跟踪,本文提出了一种基于深度学习的多旋翼无人机单目视觉目标识别跟踪方法,解决了传统的基于双目摄像机成本过高以及在复杂环境下识别准确率较低的问题。该方法基于深度学习卷积神经网络的目标检测算法,使用该算法对目标进行模型训练,将训练好的模型加载到搭载ROS的机载电脑。机载电脑外接单目摄像机,单目摄像头检测目标后,自动检测出目标在图像中的位置,通过采用一种基于坐标求差的优化算法进行目标位置准确获取,然后将目标位置信息转化为控制无人机飞行的期望速度和高度发送给飞控板,飞控板接收到机载电脑发送的跟踪指令,实现对目标物体的跟踪。试验结果验证了该方法可以很好的进行目标识别并实现目标追踪     

基于孪生区域建议网络的无人机目标跟踪算法

在无人机跟踪过程中,遮挡、光照变化、背景干扰等影响会导致跟踪目标丢失。基于SiamRPN算法提出一种无人机目标跟踪算法。通过在网络中加入空间条带池和全局上下文模块建立远程上下文关系,以适应不同的跟踪场景。同时利用改进交并比的计算方法提取目标特征,并回归精准的预测框。在UAV123数据集上的实验结果表明,相比SiamRPN、SiamFC、SAMF等算法,该算法的跟踪性能较优且具有较强的鲁棒性,尤其在背景干扰环境下,其精确率和成功率较SiamRPN算法分别提升了6.54%和11.63%。     

Maximizing Water Surface Target Localization Accuracy Under Sunlight Reflection with an Autonomous Unmanned Aerial Vehicle

Reflected sunlight can significantly impact the effectiveness of vision-based object detection and tracking algorithms, especially ones developed for an aerial platform operating over a marine environment. These algorithms often fail to detect water surface objects due to sunlight glitter or rapid course corrections of unmanned aerial vehicles (UAVs) generated by the laws of aerodynamics. In this paper, we propose a UAV path planning method that maximizes the stationary or mobile target detection likelihood during localization and tracking by minimizing the sunlight reflection influences. In order to better reduce sunlight reflection effects, an image-based sunlight reflection reception adjustment is also proposed. We validate our method using both stationary and mobile target tracking tests.     

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。     

UAV guidance using a monocular-vision sensor for aerial target tracking

Target tracking is difficult for a Unmanned Aerial Vehicle (UAV) equipped with a monocular-vision sensor because the sensor cannot measure the range between aerial target and UAV. Since the range between UAV and target is unobservable, the target position is also unknown. A measurement model of the vision sensor is proposed based on a specific image processing technique. A nonlinear adaptive observer is designed to estimate states and parameters, and the position of the target is estimated. A guidance law for target tracking and UAV maneuvers for persistent excitation condition are also proposed. To demonstrate the effectiveness of the proposed algorithms, numerical simulations are performed.     

Incremental Focus of Attention for Robust Vision-Based Tracking

We present the Incremental Focus of Attention (IFA) architecture for robust, adaptive, real-time motion tracking. IFA systems combine several visual search and vision-based tracking algorithms into a layered hierarchy. The architecture controls the transitions between layers and executes algorithms appropriate to the visual environment at hand: When conditions are good, tracking is accurate and precise; as conditions deteriorate, more robust, yet less accurate algorithms take over; when tracking is lost altogether, layers cooperate to perform a rapid search for the target and continue tracking.Implemented IFA systems are extremely robust to most common types of temporary visual disturbances. They resist minor visual perturbances and recover quickly after full occlusions, illumination changes, major distractions, and target disappearances. Analysis of the algorithm's recovery times are supported by simulation results and experiments on real data. In particular, examples show that recovery times after lost tracking depend primarily on the number of objects visually similar to the target in the field of view.     

Automatic Take Off, Tracking and Landing of a Miniature UAV on a Moving Carrier Vehicle

We present a system consisting of a miniature unmanned aerial vehicle (UAV) and a small carrier vehicle, in which the UAV is capable of autonomously starting from the moving ground vehicle, tracking it at a constant distance and landing on a platform on the carrier in motion. Our visual tracking approach differs from other methods by using low-cost, lightweight commodity consumer hardware. As main sensor we use a Wii remote infrared (IR) camera, which allows robust tracking of a pattern of IR lights in conditions without direct sunlight. The system does not need to communicate with the ground vehicle and works with an onboard 8-bit microcontroller. Nevertheless the position and orientation relative to the IR pattern is estimated at a frequency of approximately 50 Hz. This enables the UAV to fly fully autonomously, performing flight control, self-stabilisation and visual tracking of the ground vehicle. We present experiments in which our UAV performs autonomous flights with a moving ground carrier describing a circular path and where the carrier is rotating. The system provides small errors and allows for safe, autonomous indoor flights.     

引入视觉注意机制的目标跟踪方法综述

视觉跟踪在无人飞行器、移动机器人、智能监控等领域有着广泛的应用,但由于目标外观和环境的变化,以及背景干扰等因素的存在,使得复杂场景下的鲁棒实时的目标跟踪成为一项极具挑战性的任务. 视觉注意是人类视觉信息处理过程中的一项重要的心理调节机制,在视觉注意的引导下,人类能够从众多的视觉信息中快速地选择那些最重要、最有用、与当前行为最相关的感兴趣的视觉信息,特别地,人类能够快速指向感兴趣的目标,从而可以轻松地实现对目标的稳定跟踪.因此,将视觉注意机制引入到复杂场景下的目标跟踪中,有利于实现更为稳定和接近于人类认知机制的视觉跟踪算法.本文旨在对引入了视觉注意机制的目标跟踪方法进行综述. 首先,介绍了视觉注意的基本概念及其代表性的计算模型;其次,对视觉注意与跟踪的内在关系进行了阐述;然后,对引入视觉注意机制的目标跟踪方法进行归纳、总结和分类,对代表性的方法进行介绍和分析;最后,对该类方法的特点和优势进行了讨论,并对未来的研究趋势进行了展望.     

Development of a vision-based ground target detection and tracking system for a small unmanned helicopter

It is undoubted that the latest trend in the unmanned aerial vehicles (UAVs) community is towards visionbased unmanned small-scale helicopter, utilizing the maneuvering capabilities of the helicopter and the rich information of visual sensors, in order to arrive at a versatile platform for a variety of applications such as navigation, surveillance, tracking, etc. In this paper, we present the development of a visionbased ground target detection and tracking system for a small UAV helicopter. More specifically, we propose a real-time vision algorithm, based on moment invariants and two-stage pattern recognition, to achieve automatic ground target detection. In the proposed algorithm, the key geometry features of the target are extracted to detect and identify the target. Simultaneously, a Kalman filter is used to estimate and predict the position of the target, referred to as dynamic features, based on its motion model. These dynamic features are then combined with geometry features to identify the target in the second-stage of pattern recognition, when geometry features of the target change significantly due to noise and disturbance in the environment. Once the target is identified, an automatic control scheme is utilized to control the pan/tilt visual mechanism mounted on the helicopter such that the identified target is to be tracked at the center of the captured images. Experimental results based on images captured by the small-scale unmanned helicopter, SheLion, in actual flight tests demonstrate the effectiveness and robustness of the overall system.